Trichoderma sp. strain T154 was isolated from the wood of Vitis vinifera cv. Tempranillo, at Winery “Pago de Carraovejas” in Peñafiel, 41°35′51"N, 4°07′22"W (Valladolid, Spain) (Carro-Huerga et al., 2020). This strain was stored at the “Laboratorio de Diagnóstico de Plagas y Enfermedades Vegetales” (Plant and Pest Diagnostic Laboratory) (LDPEV) Universidad de León, Spain, under accession code ULET154.

The four Trichoderma strains used for morphological comparison purposes were as follows: T. lentiforme CBS 100542, Trichoderma atrobrunneum CECT 20730, Trichoderma guizhouense CECT 20731, and Trichoderma harzianum sensu stricto, which have been deposited at LDPEV under accession number ULET87 (University of Leoín, Spain). This strain (ULET87) was isolated from vineyard soils in Castilla y León region, and it has been assayed against Phaeoacremonium minimum during an in-vitro test with a significant percentage of biocontrol (Carro-Huerga et al., 2023).

Pathogens used in biocontrol assays, available at the LDPEV collection, belong to the main GTDs and most aggressive pathogens described in the literature (Martín and Cobos, 2007) in our region: Diplodia seriata (ULEP32), which is the most representative and aggressive pathogen of Botryosphaeria dieback disease, was isolated from plants with severe symptoms of xylem necrosis and shoot dieback in Castilla y León (Spain). Two of the main pathogens that cause Petri and esca diseases were also isolated: P. minimum Y038-05-3, a very aggressive isolate, was isolated from vine plants in Valle Benavente (Spain) (Martín and Martín, 2013) (stored at LDPEV under the accession number ULEA16); and finally, Phaeomoniella chlamydospora isolate Y-116-18-03c, another causal agent described as an important pathogen that causes symptoms of esca disease (Martín et al., 2012) (stored at LDPEV under the accession number ULEC21), was isolated.

The T154 isolate was isolated in a previous study (Carro-Huerga et al., 2020), carried out in the Pago de Carraovejas winery (Peñafiel, Spain). Briefly, vine bark samples were taken from various plants using pruning shears, which were disinfected with 70% ethanol between samples. These wood pieces were preserved in clean plastic bags with hermetic seals at 4°C. The bark samples were then disinfected in a 1.5% sodium hypochlorite solution for 1 min and then washed with plenty of sterile distilled water. These bark fragments were dried out for 15 min in a laminar flow chamber and cut with a sterile scalpel. Subsequently, seven wood chips (approx. 1–2 mm in diameter; approx. 0.5–1 cm in length) per plate were placed on Rose Bengal-Chloramphenicol Agar plates (Conda Laboratory, Madrid, Spain). These cultures were incubated at 25°C. After 3–5 days, mycelial growth was observed on each of the tissue pieces of wood, and those that were morphologically identified with Trichoderma were isolated and cultured on PDA plates (Sigma-Aldrich Chemie GmbH, Steinheim, Germany). The morphology of the Trichoderma strain was evaluated according to Harman and Kubicek (1998a).

For a preliminary identification of the Trichoderma isolates recovered in the present study, a strategy based on PCR amplification, nucleotide sequencing of ITS regions, and Blastn comparison to sequences in the non-redundant GenBank NCBI database (http://www.ncbi.nlm-nih.gov), using the BLASTn program (http://www.ncibi.nlm.nigh.gov/BLAST), was followed as described previously (Carro-Huerga et al., 2020).

The genome sequence from the T154 isolate was generated by Macrogen Inc. (Seoul, Korea; https://dna.macrogen.com) using an Illumina platform. Sequence assemblies were generated using Platanus Allee (v2.2.2) software (Kajitani et al., 2019). This Whole Genome Shotgun project has been deposited at DDBJ/ENA/GenBank under the accession JAZAQE000000000. The version described in this paper is version JAZAQE010000000.

Nucleotide sequences from 20 Trichoderma housekeeping (=HK) genes ( Supplementary Table S1 ) retrieved from the genome of 24 Trichoderma species (Gutiérrez et al., 2021) were used to infer a Trichoderma species tree.

The sequences of each gene from all the Trichoderma spp. analyzed were individually aligned by MUSCLE software as implemented in MEGA X (Kumar et al., 2018), and then the alignments were concatenated using the Sequence Matrix software (Vaidya et al., 2011). The resulting concatenated alignment was then subjected to maximum likelihood (ML) analysis as implemented in the program IQ-TREE version 1.6.12 (Nguyen et al., 2014). A second concatenated-partitioned tree was constructed by selecting for each gene the best-fit evolutionary nucleotide model deduced from the previous IQ-TREE analysis. Finally, both concatenated (non-partitioned and partitioned) alignments were subjected to ML analysis as implemented in IQ-TREE. Branch support was determined by bootstrap analysis using 1,000 pseudoreplicates. In addition, to assess the consistency of trees inferred from the 20 housekeeping genes, a gene concordance factor (GCF) analysis was performed as described by Minh et al. (2020) and Gutiérrez et al. (2021).

In order to support these studies, three other trees were inferred using partial amino acid sequences deduced from coding sequences of three housekeeping genes [acl1 (ATP citrate lyase), rpb2 (RNA polymerase 2nd largest subunit), and tef1 (translation elongation factor 1-alpha)], which were retrieved from different species of (Jaklitsch and Voglmayr, 2015). These sequences were aligned by MUSCLE software as implemented in MEGA X, and the trees were generated with the program IQ-TREE version 1.6.12. Branch support was assessed by a bootstrap analysis based on 1,000 pseudoreplicates.

Following the method described in Zheng et al. (2021) with some modifications, growth rate and optimal growth temperature were determined on 90-mm-diameter Petri dishes on three different culture media: potato dextrose agar (PDA), corn meal dextrose agar (CMD; 40 g of cornmeal, 20 g of glucose, 18 g of agar, 1 L of distilled water), and synthetic low nutrient agar (SNA; 1 g of KH₂PO₄, 1 g of KNO₃, 0.5 g of MgSO₄, 0.5 g of KCl, 0.2 g of glucose, 0.2 g of sucrose, 18 g of agar, 1 L of distilled water) at 25°C, 30°C, and 35°C. Plugs of 6 mm diameter were extracted from the edge of 7-day-old PDA plates and placed approximately 1 cm from the border of the Petri dishes. Colony radii were measured after 24 h, 48 h, 72 h, 96 h, and 7 days. The time point when mycelium completely covered the surface of the plate was also recorded during this assay. Furthermore, the morphological characters of the colonies, such as their appearance, color, and spore production, were recorded at the same time. The development of T154 was also verified at 37.5°C and 40°C, to determine the maximum temperature at which it can grow.

The T154 isolate was cultured in PDA for the evaluation of the microscopic morphology, and it was incubated at 25°C for a 72–96-h period. Pictures were taken with a Nikon Eclipse E600 microscope connected to a Nikon DS-Fi3 digital camera.

Spore production was evaluated with four replicates of each of the following isolates, T154, T. harzianum, T. atrobrunneum, T. guizhouense, and T. lentiforme inoculated in PDA medium and incubated for 7 days at 25°C. Spores were collected by washing the Petri dish with distilled water and then filtering the spore suspension through a filter cloth. The spores were counted with a Neubauer chamber.

The antagonistic capacity of the T154 isolate was evaluated in vitro against the grapevine trunk disease pathogens P. minimum, P. chlamydospora, and D. seriata by performing dual culture tests. First, mycelial plugs (6 mm in diameter) of P. minimum and P. chlamydospora were obtained from the edge of 7-day-old PDA culture plates grown at 25°C. The plugs of these two pathogens were incubated for 14 days at 25°C in PDA to give them an advantage over Trichoderma, due to their slow growth rate. Subsequently, mycelial plugs (6 mm in diameter) of T154 were placed next to the pathogens, at a distance of 55 mm, from the edge of the cultures on 7-day-old PDA plates grown at 25°C. In the case of D. seriata, both mycelial plugs were placed at the same time on the plate at a distance of 55 mm since they showed a similar speed of development. Phaeoacremonium minimum, P. chlamydospora, and D. seriata were also incubated individually without the presence of Trichoderma under the same environmental conditions as the controls.

Dual cultures were incubated at 12°C in the dark to simulate the behavior of the T154–pathogen interaction at the time of pruning in the field, in which the pathogens have a greater possibility of penetrating the vine plant. Each control and Trichoderma/pathogen combination was replicated four times. Pathogen growth diameters were measured over a period of 30 to 40 days, depending on the development time required for each pathogen. The inhibition percentage caused by Trichoderma was calculated with the following equation (Úrbez-Torres et al., 2020):

where D1 is the diameter of the pathogen mycelium grown in the presence of Trichoderma and D2 is the diameter of the pathogen mycelium grown alone in the control plate.

An evaluation of the resistance of T. carraovejensis against five pesticides and three biostimulants (commonly used in commercial wineries) was carried out in this experiment. This assay helps us to preliminarily identify which products could affect the T. carraovejensis strain and reduce its efficacy as a biological control agent.

The test was carried out in 90 mm diameter Petri dishes with PDA medium that was amended with the concentrations of each product indicated in Table 1 . Pesticides and biostimulants were added to the PDA-melted medium at 45°C, the medium was homogenized in a magnetic agitator at 500 rpm, and 15 ml of the solution was poured into each Petri dish. Subsequently, inoculation of each plate was carried out by placing a 6-mm diameter mycelium plug from T. carraovejensis of a 7-day-old PDA culture. PDA plates without adding any product were used as control. Four repetitions per treatment were made. Plates were kept at room temperature (24°C) and measurements of the diameters of each treatment were made on the 2nd, 4th, and 7th days after inoculation.

All of the tests that were carried out were analyzed using IBM SPSS^® Statistics 21 (IBM Corp., Armonk, NY, USA). This software was used for the statistical analyses as follows: first, the Shapiro–Wilk test was used to check if there was normal distribution, then the homogeneity of variances was evaluated using Levene’s test, and one-way ANOVA was carried out to determine if there were significant differences. A post-hoc test (Duncan, p< 0.05) was performed to establish differences between groups.

The 25 Trichoderma species included in the phylogenetic analysis belong to 11 previously described lineages (Kubicek et al., 2019; Gutiérrez et al., 2021). To establish the position of the T154 isolate, we inferred a species tree based on a maximum likelihood analysis of concatenated alignments of 20 housekeeping genes that were retrieved in previous works from Trichoderma genome sequences corresponding to ex-type strains for each of the species used (Gutiérrez et al., 2021) as well as from the genome sequence of the T154 strain. In the resulting tree ( Figure 1 ), our species of interest, T154, represents an individual branch within the Harzianum/Virens clade, with a bootstrap value of 100 and a GCF of 6. Therefore, these data support the assignment of the T154 isolate to a new species, which has been named T. carraovejensis. Furthermore, the topology of the phylogenetic tree was largely consistent with previously reported multispecies phylogenies for other Trichoderma species combinations (Jaklitsch and Voglmayr, 2015; Kubicek et al., 2019; Gutiérrez et al., 2021).

Three of the genetic markers (acl1, tef1, and rpb2) used in the study indicated above were analyzed separately. Thus, phylogenetic analyses of the acl1, tef1, and rpb2 partial genes from a wider isolate representation resulted in individual trees that are consistent ( Supplementary Figures S1 - S3 ) with the main tree of the 20 housekeeping genes ( Figure 1 ). In the three individual trees, the bootstrap values obtained on the T. carraovejensis branch were ≥85%.

The results on the effect of different temperatures (25°C–40°C) on T. carraovejensis growth are summarized in Figure 2 . The optimal growth temperature was estimated at 30°C for PDA and CMD culture media after 72 h from inoculation. For SNA, the optimal growth temperature varies between 25°C and 30°C, with no significant differences between these two temperatures. To conclude, the best media for T. carraovejensis growth are PDA and CMD, with no significant differences between them at 25°C, 30°C, and 35°C. However, there were significant growth differences between the first two media and SNA regardless of the temperature.

Furthermore, there was a significantly lower development of T. carraovejensis at 37.5°C in all media, as well as complete inhibition of fungal development at 40°C ( Figure 2 ).

Further experiments were carried out in order to compare T. carraovejensis with the different species that were closely related in the phylogenetic trees: Trichoderma harzianum ULET87, T. atrobrunneum CECT 20730, T. guizhouense CECT 20731, and T. lentiforme CBS 100542. First, the growth at 25°C, 30°C, and 35°C was assessed in different culture media.

After 72 h of inoculation at 25°C on PDA, the colony of T. carraovejensis reaches a radius of 57.5 ± 1 mm, showing significant differences only with T. harzianum. Under the same conditions in CMD and SNA, T. carraovejensis growth does not differ significantly from the other species. A radius of 62 ± 1 mm in CMD and 27 ± 1 mm in SNA was reached. At 30°C after 72 h of growth from inoculation, T. carraovejensis demonstrated great growth rates in the different culture media used, standing out in the case of the PDA (74.5 ± 1 mm) and SNA (32.5 ± 1 mm) media above the rest of the species and in the case of the CMD (75 ± 1 mm) obtaining the highest growth rates, although with less difference with respect to the other species. In the case of PDA and SNA, a significant difference was observed in the development of T. carraovejensis with respect to the rest of the species. Under these conditions, T. carraovejensis completely covered the plate in CMD medium. At 35°C after 72 h of inoculation, T. carraovejensis reached a radius of 33.25 ± 1 mm in PDA, which is significantly higher than those reached by T. harzianum (25.25 ± 1 mm), T. atrobrunneum (14.25 ± 1 mm), and T. guizhouense (16.75 ± 1 mm) but identical to that reached by T. lentiforme (33.25 ± 1 mm). In CMD, T. carraovejensis also reaches one of the highest radii (35 ± 1 mm), being significantly higher than those reached by T. harzianum (17 ± 1 mm) and T. atrobrunneum (7 ± 1 mm). In the case of the SNA medium, T. carraovejensis reached a radius (13.5 ± 1 mm) that was significantly higher than those reached by T. atrobrunneum (7 ± 1 mm), T. guizhouense (8.5 ± 1 mm), and T. lentiforme (10.75 ± 1 mm), but significantly lower than that reached by T. harzianum (17 ± 1 mm). For most of the strains analyzed, a drastic decrease in growth was observed in all the media used when the growth was assessed at 35°C ( Supplementary Figure S4 ).

Finally, a comparison between conidia production was performed in this experiment. An evaluation of the production of conidia was done 7 days after inoculation on PDA medium at 30°C. Trichoderma carraovejensis was able to produce 1.13 × 10⁹ conidia/ml, a higher value than the other species used for comparison: 1.75 × 10⁸ conidia/ml of T. harzianum, 3.59 × 10⁸ conidia/ml for T. atrobrunneum, 1.01 × 10⁹ conidia/ml of T. guizhouense, and 2.19 × 10⁸ conidia/ml for T. lentiforme, which represent 84.51%, 68.23%, 10.62%, and 80.62% less than T. carraovejensis, respectively ( Figure 3 ).

Morphological characterization for the new species T. carraovejensis at three different temperatures showed different features ( Figure 4 ). First, T. carraovejensis varied depending on the growth temperature. Development at 25°C gave rise to the appearance of yellow-green pigments. Second, at 30°C, the pigments were greenish tones, and finally at 35°C, yellow-orange tones could be observed at the base of the culture medium ( Figure 4 ).

Thus, according to the results, the most useful culture medium for morphological comparison was PDA. At 30°C, all the compared species developed a growth pattern in concentric circles, with T. carraovejensis having the largest radius. At this temperature, T. carraovejensis began its development with a dense white mycelium. The first pigments to appear in the plate at 48 h were yellow tones, shortly after transforming to light greenish tones with subsequent darkening as the number of conidia in the plate increased. One of the main differences in comparison to the rest of the Trichoderma strains was observed after 7 days of growth when T. carraovejensis was the only one that no longer had areas with white mycelium. The entire plate was covered with conidia. This fact does not occur with the other species, which after 7 days continue to present alternate white mycelium zones between the conidia generation zones ( Figure 5 ).

Comparison of morphological characterization was based on an in-vitro experiment at 35°C, which is the best range of temperature to differentiate T. carraovejensis from its closest Trichoderma-related species ( Supplementary Figure S5 ). Trichoderma harzianum sensu stricto, T. atrobrunneum, and T. guizhouense were not able to colonize more than half of the Petri dish plate in PDA, so this is a morphological criterion that can discriminate close species as different ones. If we compare T. carraovejensis and T. lentiforme, the whole plate was covered after 7 days in PDA at 35°C, and similar concentrical rings could be visualized and green pale olive colors were identified. However, from the point of inoculation, T. lentiforme had a yellowish and pale white color, which was also observed at the border of the plate where white color was identified. However, T. carraovejensis presented a homogeneous green pale color except at the point of inoculation where a strong green color was identified; possibly, it was due to the high concentration of conidia in comparison to the rest of Trichoderma species as confirmed in Figure 3 .

T. carraovejensis—G. Carro-Huerga, L. Zanfaño, S. Gutiérrez, P.A. Casquero Luelmo. sp. nov. ( Figure 6 ).

Etymology —Referring to the winery where the isolate was obtained: “Pago de Carraovejas”.

Typus —Spain, Castilla y León, Valladolid, Peñafiel, grapevine bark of vine plant (Vitis vinifera), G. Carro-Huerga (GenBank accession number JAZAQE000000000).

Fast growing colonies. At the beginning of the development of the fungus, white cottony pustules were formed, which subsequently sporulated, taking on greenish colors and a granular texture. Pallid green reverse. Mycelium composed of branched, septate and hyaline hyphae. The conidiophores presented paired lateral branches and lageniform to utriform phialides, elongated in shape and broader at the base, appeared individually or in groups of two or three, (10.6–)12.0–12.2(–16.2) × (2.1–)2.8–2.9(–3.0) μm, length/width ratio. The globose-shaped conidia with smooth edges showed greenish tones, 2.1–2.5 × 2.6–2.8 μm, length/width ratio ( Figure 6 ).

Culture characteristics— Optimal growth temperature at 30°C. Colony radius on CMD after 72 h of growth: 62 ± 1 mm at 25°C, 75 ± 1 mm at 30°C, and 35 ± 1 mm at 35°C, covering the plate after 3 days at 30°C. Translucent mycelium at 48 h, appearance of the first yellow-green tones at 96 h with the formation of spores. Dense and greenish mycelium at 7 days with greater spore production at the edges of the plate. Colony radius on PDA after 72 h: 57.5 ± 1 mm at 25°C, 74.25 ± 1 mm at 30°C, and 33.25 ± 1 mm at 35°C, covering the plate after 4 days at 25°C and 30°C. Formation of dense mycelium in concentric circles. Appearance of the first spores at 48 h around the sowing disc. Green pigments that increase in intensity as the days pass since inoculation, which corresponds to an increase in the number of spores produced. Colony radius on SNA after 72 h: 27 ± 1 mm at 25°C, 32.5 ± 1 mm at 30°C, and 13.5 ± 1 mm at 35°C, covering the plate after 7 days at 25°C and 30°C. Formation of a sparse mycelium with greenish pigments after 48 h. The production of spores in PDA at 7 days was 1.13 × 10⁹ spores/ml.

Notes —In this study, T. carraovejensis was isolated from grapevine bark. Trichoderma carraovejensis was differentiated from other species by phylogenetic analysis of the sequences of 20 housekeeping genes from 24 different Trichoderma species and by morphological comparison with T. harzianum, T. atrobrunneum, T. guizhouense, and T. lentiforme, the species closest to it in the generated phylogenetic tree.

The levels of antagonism of T. carraovejensis against the GTD pathogens P. minimum and P. chlamydospora were evaluated after 38 days of growth and for D. seriata after 31 days. This was due to the slow development of the isolates at 12°C. The mean percentages of radial growth inhibition (RI) for P. minimum, P. chlamydospora, and D. seriata were 21.04%, 15.34%, and 34.08%, respectively. Trichoderma carraovejensis exhibited the best inhibition values against the pathogen D. seriata when assays were carried out at 12°C ( Figure 7 ).

Trichoderma was able to stop the development of P. minimum but without colonizing it. In the case of P. chlamydospora, T. carraovejensis stopped the growth of the pathogen, and the biocontrol agent was able to colonize the pathogenic fungus, growing and producing spores on it. As for D. seriata, a clear arrest of the development of the pathogen and an overgrowth of Trichoderma on it were observed.

Assays performed with T. carraovejensis grown in the presence of different products showed significant differences (p< 0.05) among pesticides and biostimulants. The colony diameter in the control plate (=Trichoderma growing alone in PDA medium) reached the highest value, 72.25 ± 1 mm, and this value was significantly higher in comparison to the rest of the tested products. Trichoderma carraovejensis demonstrated the greatest compatibility with the 80% sulfur-based product (62.75 ± 1 mm) and differed significantly to 98.5% sulfur (51.00 ± 1 mm) and the products described in Table 1 . Trichoderma reached a diameter of 30.88 ± 1 mm in combination with the product based on amino acids and iron. Very similar values were obtained with the copper-based products, with no significant differences between Cu 35% (14.75 ± 1 mm) and Cu5.5 + Cu2.8 (11.88 ± 1 mm) and between the latter and Cu+Mn+Zn (10.00 ± 1 mm). No significant differences were also observed between Cu+Mn+Zn and 72% sulfur (7.13 ± 1 mm). Cinnamon showed the lowest compatibility with T. carraovejensis, inhibiting its growth ( Figures 8 , 9 ).