Laboratory assessment of novel endophytic Trichoderma-based bioformulations for the biological control of sorghum leaf spot and stalk rot diseases

Abstract

Stalk rot and leaf spot, caused by Fusarium verticillioides and Curvularia lunata, respectively, are among the most destructive seed-borne fungal diseases of sorghum, resulting in substantial yield losses. In the present study, two novel endophytic strains, Trichoderma asperellum SEPA11A and Trichoderma harzianum SEPA11B, were isolated from healthy sorghum seeds and evaluated for their biocontrol potential and growth-promoting effects. Out of 39 Trichoderma isolates screened, the isolates 35 and 3 exhibited the highest antagonistic activity in dual culture assays, inhibiting the mycelial growth of F. verticillioides by 68.88% and 66.67%, and C. lunata by 57.77% and 59.30%, respectively. Both strains were identified based on morphological characteristics and internal transcribed spacer (ITS) rDNA sequencing and have been deposited in GenBank under accession numbers LC866760.1 and LC866759.1, and coded as SEPA11B and SEPA11A respectively. Gas chromatography–mass spectrometry (GC–MS) analysis of culture filtrates revealed 55 bioactive metabolites, while scanning electron microscopy demonstrated severe ultrastructural damage to pathogen hyphae. A granular formulation of SEPA11B and SEPA11A maintained high viability for up to 14 months under storage. These results collectively indicate that SEPA11B and SEPA11A represent promising, sustainable, and cost-effective alternatives to chemical fungicides for integrated management of stalk rot and leaf spot diseases, with the added as a potential benefit of enhancing sorghum growth and productivity. This unique trait makes seed-derived endophytic Trichoderma isolates promising for effective, persistent biofungicide development. Adapted to the seed-soil interface, they can colonize the rhizosphere, outcompete pathogens, and remain active under field conditions. Thus, they are strong candidates for sustainable biofungicides, pending greenhouse/field validation and large-scale production optimization.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-54927-w.

Introduction

Sorghum (Sorghum bicolor L.) is a globally important cereal crop, ranking fifth among cereal grains worldwide¹. Beyond its critical role in ensuring food security, sorghum is also a valuable resource for animal feed and biodiesel production. The crop is primarily cultivated under dry and semi-arid conditions and is widely grown across Africa, Asia, and the Americas. Africa alone contributes approximately 20 million tons to the global annual harvest. However, the increasing prevalence and severity of sorghum diseases are not occurring in a vacuum; they are being intensified by the growing pressures of climate change. Environmental stressors—such as prolonged drought and extreme temperature fluctuations—can weaken the plant’s natural defense mechanisms, making it more vulnerable to infection. This climate-induced vulnerability creates opportunities for pathogens to thrive, transforming what were once minor diseases into major economic threats². According to FAOSTAT³, Egypt cultivates 150,000 hectares of sorghum, producing approximately 780,000 tons yearly. Under Egypt Vision 2030, expansion of drought-tolerant crops such as sorghum is a national priority to address climate change challenges, achieve food security, and promote sustainable agriculture. Data from the Foreign Agricultural Service (2025) indicate that Asyut, Sohag, and Fayoum governorates collectively account for more than 80% of Egypt’s sorghum production.

Sorghum is susceptible to more than 50 diseases, the majority of which are seed-borne, with fungal pathogens representing the most significant constraint to production⁴. Seed-borne fungal diseases significantly reduce sorghum yield and quality, impair germplasm and seed production, and limit domestic and international trade⁵. Seed treatment is considered the most effective approach for controlling these pathogens⁶. Seed treatment is considered among the most effective approaches for managing seed-borne pathogens, as it protects germinating seedlings during their most vulnerable stage. This localized protection enhances germination, promotes early plant vigor, and reduces the need for extensive fungicide applications later in crop development^7–9. The reported yield loss of 12–55% for Curvularia lunata (Curvularia leaf spot) in sorghum reflects region-specific conditions, particularly in tropical and subtropical areas with high humidity and rainfall, rather than a uniform global average^10–12. In Egypt, C. lunata has been identified as one of the major fungi affecting sorghum grain genotypes under both storage and field conditions^13,14. Recently, it has emerged as a significant pathogen responsible for sorghum leaf spot and grain mold, adversely impacting both grain quality and yield¹⁵. Fusarium stalk rot, caused by several Fusarium spp., is another major disease of sorghum. The pathogen colonizes roots and stalk tissues, causing vascular discoloration, cortical tissue degradation, and impaired water and nutrient translocation, often resulting in plant lodging under severe infection¹⁶. Fusarium mycotoxins, particularly fumonisins, zearalenone, and deoxynivalenol (DON), frequently contaminate sorghum grains, posing significant health risks to both humans and livestock. Humans are exposed primarily through consumption of contaminated sorghum-based food products, while livestock may be affected via contaminated feed, leading to chronic health issues such as esophageal cancer, neural tube defects, immune suppression, and reduced productivity^17,18.

Chemical fungicides remain a primary management strategy but raise concerns regarding environmental contamination, non-target effects on beneficial microbiota, and risks to human and animal health¹⁹. Additionally, the emergence of fungicide-resistant pathogen populations has intensified the search for eco-friendly alternatives²⁰. Biological control represents a sustainable, non-toxic, and cost-effective alternative, often reducing management costs by up to 50% compared with chemical fungicides^21,22. Moreover, biocontrol agents (BCAs) enhance plant growth, induce systemic resistance, and improve yield²³. Numerous studies have demonstrated the efficacy of biocontrol agents in seed treatment against a wide range of seed-borne fungi²⁴.

Among biocontrol agents, Trichoderma species are the most effective antagonists against several plant pathogens, including fungi. Trichoderma functions as a biocontrol agent through two primary mechanisms: Direct antagonism and induced systemic resistance (ISR). Direct antagonism occurs when the fungus physically or chemically attacks pathogens via mycoparasitism (coiling and enzyme degradation), antibiosis (toxin production), and competition for nutrients. In contrast, ISR involves colonization of plant tissues that primes the plant’s own defenses, enhancing production of phenolics, strengthening cell walls, and activating defense pathways without direct contact with the pathogen^25,26. Trichoderma spp. rapidly colonizes the rhizosphere, forming a protective barrier against pathogen invasion²⁷, and have been shown to be effective against various pathogens, including Sclerotinia sclerotiorum, S. rolfsii, Macrophomina phaseolina, Rhizoctonia solani, Pythium spp., Phytophthora spp., Fusarium spp. and Curvularia spp.^28–35. Additionally, Trichoderma produces secondary metabolites that contribute to pathogen suppression, plant growth promotion, systemic resistance induction, and control pest³⁶. Trichoderma biocontrol agents are commonly produced using cost-effective solid-state (SSF) and liquid (LSF) fermentation, with carrier materials such as wheat bran, semolina, or vermiculite used to stabilize the formulations. Solid-state methods often provide higher viability, whereas liquid methods allow efficient large-scale production. Additionally, new technologies, including encapsulation and the use of carriers, enhance shelf life and improve field efficacy^37,38. Endophytic Trichoderma isolated from seeds represents a novel source compared to rhizospheric isolates due to its unique ecological niche and lifestyle. Unlike soil-dwelling rhizospheric strains, which are generally opportunistic, seed endophytes are specialized to colonize internal plant tissues, allowing them to persist within the host, transmit vertically through seeds, and provide systemic protection against pathogens from the earliest stages of plant development. This internal colonization often confers enhanced tolerance to environmental stresses such as drought, heat, and microbial competition, and can result in greater antagonistic activity against seed-borne pathogens^39,40. The exploitation of seed-derived endophytic Trichoderma is therefore particularly promising for the development of biofungicides, as these strains are already adapted to the internal seed environment, ensuring effective colonization, long-term persistence, and consistent biocontrol activity when applied to crops.

Commercial formulations of T. harzianum and T. viride are currently used against more than 70 soil-borne and 18 foliar pathogens affecting over 87 crops³⁹. The success of Trichoderma-based products is attributed to their rhizosphere competence, broad-spectrum activity, and compatibility with other BCAs⁴¹. The shelf life of formulations varies with storage conditions and carrier material, ranging from 3 to 4 months in kaolin-based carriers to over 18 months in coffee-husk-based formulations⁴². The literature highlights a need for more comprehensive and current studies on seed-borne pathogens specifically affecting sorghum in Egypt. Furthermore, there is limited information on locally adapted Trichoderma strains for effective biocontrol of these pathogens. This study addresses these gaps by isolating, characterizing, and evaluating endophytic Trichoderma strains from Egyptian sorghum seeds, providing both scientific insights and practical strategies for managing seed-borne fungal diseases⁴³.

This study addresses key research gaps by focusing on seed-derived endophytic Trichoderma strains from sorghum in Egypt, which offer unique advantages over conventional rhizospheric strains due to their internal colonization, vertical transmission, and enhanced antagonism against seed-borne pathogens. While previous research has largely relied on in vitro antagonism without linking metabolite profiles to biocontrol function, and has seldom developed granular formulations from locally adapted strains, this study combines molecular characterization, metabolite profiling, and formulation development to create a stable, targeted granular biofungicide for sustainable management of sorghum leaf spot and stalk rot.

Results

Distribution of sorghum endophytic seed-borne mycobiota

Seed-borne fungal endophytes were detected and identified from healthy-looking sorghum seed samples collected across six major sorghum-growing governorates in Egypt during the 2023 cropping season. A total of 22 composite seed samples were analyzed using both the agar plate (AP) and deep-freezing blotter (DFB) techniques. These analyses revealed 16 fungal genera encompassing 22 endophytic species (Figs. 1 and 2). Among the recovered taxa, Alternaria alternata, Penicillium spp., and Aspergillus flavus were the most frequently detected endophytic fungi across all sorghum-growing regions, followed by Aspergillus niger, Bipolaris spp., Cladosporium spp., Curvularia lunata, Fusarium verticillioides, and Trichoderma spp. Less frequent species included Aspergillus terreus, Epicoccum nigrum, and Fusarium incarnatum. Detailed frequency and incidence data are summarized in Figs. (1 and 2). In contrast, Aspergillus terreus, Cephalosporium spp., Epicoccum nigrum, and Fusarium incarnatum (synonym: F. semitectum) were the least frequently detected (4.55 each with the AP), while Aspergillus glaucus, Chaetomium sp., Fusarium equiseti, Nigrospora spp., and Ulocladium spp. recorded the lowest frequency with DFB. Notably, E. nigrum was exclusively detected with AP, whereas A. glaucus and Ulocladium sp. were exclusively detected with DFB (Fig. 1). Several isolates of the pathogenic fungus C. lunata were consistently detected, with a frequency of 54.55% using both AP and DFB techniques. In contrast, isolates of F. verticillioides exhibited frequencies of 54.55% with AP and 50.00% with DFB. Cephalosporium spp. exhibited a comparatively lower frequency rate, 4.55% (AP) and 13.63% (DFB). Among potential biocontrol fungi, Trichoderma spp. was the only taxon consistently detected by both AP and DFB methods, with a frequency of 40.91% (Fig. 2). Differences in fungal detection between the agar plate (AP) and deep-freezing blotter (DFB) methods were observed. The AP technique generally detected a broader range of fungi, whereas the DFB method facilitated the identification of certain slower-growing pathogens. A detailed explanation of the methodological sensitivity differences is provided in the discussion section.

Fig. 1.

Illustrating the frequency distribution (mean ± SE) of 22 endophytic seed-borne fungal taxa isolated from sorghum seeds collected from six sorghum-growing governorates in Egypt during the 2023 cropping season. Frequencies were determined using both the deep-freezing blotter (DFB) and agar plate (AP) techniques.

Fig. 2.

Depicting the mean incidence (% of seeds infected) (mean ± SE) of 22 endophytic seed-borne fungal taxa recovered from sorghum seeds collected from six sorghum-growing governorates in Egypt during the 2023 cropping season. Incidences were determined using both the deep-freezing blotter (DFB) and agar plate (AP) techniques.

Dual culture antagonism assay

The antagonistic potential of 39 isolates of Trichoderma spp. against F. verticillioides (SOR2) and C. lunata (SOR7) was evaluated under in vitro conditions. Results revealed that most displayed marked inhibitory effects on mycelial growth, with growth reduction percentages ranging between 35.53% and 73.33% (Table 1).

Table 1.

Effect of Trichoderma spp. isolates (TI) on radial growth inhibition (%) of Fusarium verticillioides (SOR2) and Curvularia lunata (SOR7) in dual culture assays.

TI	Growth reduction (%)		AG	TI	Growth reduction (%)		AG
	FV-SOR2	CL-SOR7			FV-SOR2	CL-SOR7
1	53.32e-k	51.20 g-i	1	21	55.60 d-g	50.40 hi	1
2	55.55d-g	52.60 e-h	1	22	55.73 d-f	53.47 d-h	1
3	66.67ab	59.30 b	1	23	55.93 d-f	48.12 ij	1
4	52.53f-l	52.59 e-h	1	24	57.27 de	53.31 e-h	1
5	46.67mn	51.23 g-i	1	25	51.20 g-l	53.36 e-h	1
6	54.87e-j	55.57 c-e	2	26	48.67 l-n	57.80 bc	1
7	51.20 g-l	52.57 e-h	1	27	55.43 e-g	54.96 c-f	1
8	55.60d-g	50.47 hi	1	28	53.30 e-k	55.63 c-e	1
9	50.37 j-m	52.59 e-h	1	29	44.57 n	44.50 k	1
10	66.66 ab	48.20ij	1	30	60.00 cd	59.70 b	1
11	55.60 d-g	52.61 e-h	1	31	48.90 k-n	53.31 e-h	1
12	57.75 de	51.13 g-i	1	32	55.43 e-g	52.61 e-h	1
13	54.80 e-j	55.67 c-e	1	33	62.50 bc	51.17 g-i	1
14	48.88 k-n	35.53 l	1	34	54.81 e-j	54.17 d-g	1
15	55.57 d-g	51.88 f-h	1	35	68.88 a	57.77 bc	1
16	53.33 e-k	73.33 a	1	36	50.50 h-m	46.66 jk	1
17	54.90 e-i	54.96 c-f	1	37	54.07 e-j	57.07 b-d	1
18	66.70 ab	46.67 jk	1	38	54.97 e-h	50.37 hi	1
19	54.90 e-i	51.27 g-i	1	39	60.03 cd	55.55 c-e	1
20	50.43 i-m	55.67 c-e	1
P value	< 0.0001

FV-SOR2 = Fusarium verticillioides sorghum isolate 2, CL-SOR7 = Curvularia lunata sorghum isolate 7, AG = Antagonism grade. Mean values followed by different letters within the same column are significantly different according to Tukey’s test at P ≤ 0.05 (n = 3).

Among these, isolates 3 and 35 of Trichoderma spp. exhibited consistently high antifungal activity, reducing the radial growth of F. verticillioides by 66.67% and 68.88%, and C. lunata by 59.30% and 57.77%, respectively. Trichoderma isolate 10 followed closely, achieving 66.66% inhibition against F. verticillioides (Table 1; Fig. 3).

Fig. 3.

In vitro anti-fungal activities of Trichoderma isolate 3 and 35 against Fusarium verticillioides (SOR2) and Curvularia lunata (SOR7). A = Only F. verticillioides (SOR2) control; B = Only Trichoderma isolate 3 control; C = Trichoderma isolate 3 + F. verticillioides; D = Trichoderma isolate 35 + F. verticillioides; E = Only C. lunata (SOR7) control; F = Trichoderma isolate 35 control; G = Trichoderma isolate 3 + C. lunata (SOR7); H = Trichoderma isolate 35 + C. lunata (SOR7).

Upon extended incubation for an additional eight days, both Trichoderma isolate 3 and Trichoderma isolate 35 completely overgrew and suppressed pathogen colonies. Light microscopy observations revealed pronounced mycoparasitism, characterized by fragmented, vacuolated, and disrupted pathogen hyphae compared to the untreated controls. Based on their superior antagonistic performance, isolates Trichoderma isolate 3 and Trichoderma isolate 35 were selected for subsequent morphological, molecular, and biochemical characterization.

Cultural and morphological characteristics of Trichoderma spp.

On PDA, Trichoderma isolates 3 was culturally and morphologically identified as T. asperellum depending on that it produced colonies reaching 50 mm in diameter with dark green conidial regions. Microscopic examination revealed a well-developed main axis with several lateral branches. Conidiogenous cells (phialides) were arranged in whorls of two to four, measuring 7–14 μm long and 3–4.5 μm wide. Conidia were subglobose to ovoid measuring 3.5–4.6 μm, and chlamydospores were not observed (Fig. 4A).

Fig. 4.

A, Cultural and morphological characteristics of Trichoderma isolate 3 as T. asperellum; A, colony observed on a PDA plate after 5 days of incubation at 25 ± 2 °C; 1–3, Micrographs showing conidiophores and phialides; 4, conidia. B, Growth and morphological characteristics of Trichoderma isolate 35 as T. harzianum SEP11B. 1, Colony observed on a PDA plate after 5 days of incubation at 25 ± 2 °C; 2, Micrographs illustrating conidiophores and phialides; 3, conidia; 4, chlamydospores. Bars = 10 μm.

The Trichoderma isolates 35 was identified as T. harzianum similarly for it formed colonies up to 55 mm in diameter with a gray-green conidial region. Microscopically, short main conidiophores were observed, each bearing 3–5 divergent ampulliform phialides measuring 6–7.5 μm in length and 3.2–3.5 μm in width. The conidia were subglobose (2.5–3.5 μm), and both terminal and intercalary chlamydospores were present (Fig. 4B).

Molecular confirmation of T. asperellum and T. harzianum

The ITS region of both isolates was amplified and sequenced. BLASTn analysis confirmed that Trichoderma isolates 3 belongs to T. asperellum (100% identity and 100% query coverage) and was deposited in GenBank under accession number LC866760.1 and namely SEPA11A. While the Trichoderma isolates 35 was molecular confirmed as T. harzianum with 98.78–99.82% sequence identity (65–100% query coverage) and deposited in GenBank under accession number LC866759.1 and its namely code is SEPA11A.

The top ten GenBank matches for each isolate are summarized in Table S1, while a maximum likelihood phylogenetic tree illustrating their relationships with reference sequences is presented in Fig. 5. For further confirmation of species identity, phylogeny analysis was conducted using MEGA12 software employing the maximum likelihood algorithm with 1,000 bootstrap replicates (Fig. 5). Representative Trichoderma species were included as reference sequences, and two species of Protocrea were used as outgroups. The resulting phylogenetic tree revealed that T. asperellum isolate SEPA11A clustered closely with T. asperellum isolates Tasp47 (MT065826) and TV-3 (KX538814), supported by a bootstrap value of 100 (Table S1). Similarly, T. harzianum isolate SEPA11B clustered with T. harzianum isolates Tr09-4 (PQ895230) and PWN6 (MW789612), with bootstrap support values of 97 and 90, respectively (Fig. 5). These results provide strong evidence for the accurate species-level identification of both SEPA11A and SEPA11B.

Fig. 5.

Maximum likelihood phylogenetic tree based on ITS region sequences illustrates the relationship between T. asperellum SEPA11A, T. harzianum SEPA11B, and other related Trichoderma species. The tree was constructed using MEGA12 software with 1,000 bootstrap replicates. Protocrea species were included as the outgroup to root the tree. Numbers on the tree branches indicate bootstrap support values, while the scale bar represents a phylogenetic distance of 0.02 nucleotide substitutions per site.

Scanning electron microscopy (SEM)

To gain deeper insights into the antagonistic mechanisms of T. asperellum SEPA11A and T. harzianum SEPA11B against the target pathogens, scanning electron microscopy (SEM) was employed. SEM micrographs (Figs. 6 and 7) clearly revealed the mycoparasitic interactions between the Trichoderma isolates and the pathogens in dual culture assays. The hyphae of Trichoderma were observed coiling around and penetrating the hyphae of C. lunata (SOR7) and F. verticillioides (SOR2), leading to the collapse and deformation of their conidia and mycelia. In contrast, untreated pathogen cultures displayed normal, intact hyphal and conidial structures (Figs. 6 and 7).

Fig. 6.

Scanning electron micrographs showing the mycoparasitic interaction of T. asperellum SEPA11A and T. harzianum SEPA11B with Curvularia lunata (SOR7) in dual culture. A, Untreated C. lunata showing intact hyphae, short conidiophores (arrowheads) and conidia (arrow); B, Untreated Trichoderma showing characteristic hyphae, branched conidiophores and small conidia (arrows); C & D, Trichoderma conidiophores (arrows) growing over the conidia of C. lunata (arrowheads); E, Trichoderma conidiophores (arrow) in close association with collapsed C. lunata conidia (arrowhead); F–H, Mycoparasitism process, showing Trichoderma hyphae (arrows) coiling around and penetrating C. lunata hyphae (arrowheads), resulting in hyphal collapse. Scale bar = 10 μm.

Fig. 7.

Scanning electron micrographs showing the mycoparasitic activity of T. asperellum SEPA11A and T. harzianum SEPA11B against Fusarium verticillioides (SOR2) in dual culture. A, Untreated F. verticillioides with normal hyphae (arrows) and conidia (arrowhead); B, Untreated Trichoderma hyphae and conidia (arrows); C–E, Mycoparasitism process, where Trichoderma hyphae (arrows) coil around and overgrow the hyphae of F. verticillioides (arrowheads), causing collapse and deformation of pathogen hyphae. Scale bar = 10 μm.

GC-MS analysis of cell-free cultural filtrates of the bioagents

The culture filtrates of Trichoderma isolates SEPA11A and SEPA11B were analyzed using GC-MS to determine their metabolite profiles. A total of 55 bioactive compounds were detected, with relative abundances ranging from 1.98 to 23.35% for SEPA11A and from 3.8 to 27.28% for SEPA11B (Table 2; Fig. 8). Of these, 47 compounds were unique to SEPA11A, 22 were unique to SEPA11B, and 14 were shared by both isolates (Table 2).

Table 2.

Bioactive metabolite compounds identified by GC-MS in the cultural filtrates of T. asperellum SEPA11A and T. harzianum SEPA11B.

Detected Compounds	Formula	MW	SEPA11A		SEPA11B		Biological activity	Chemical class
			Area %	RT (min)	Area %	RT (min)
(S)-(+)−2-Amino-3-methyl-1-butanol	C5H13NO	103	23.35	7.76	–	–	Antifungal44	Amino alcohols
[1aR-(1aà,2à,5á,5aá,6á,8aà,9à,10aà)]-Hexadecanoic acid, 1-(hydroxymethyl)−1,2-ethanediyl ester	C35H68O5	568	2.3	45.8	–	–	–	Fatty acid ester
14-Octadecenoic acid, methyl ester	C19H36O2	296	2.00	3.16	–	–	Antifungal45	Fatty acid
1-Butene, 1-(methylthio)-, (E)-	C5H10S	102	2.93	7.46	–	–	Antifungal46	Thioenol ethers
1-Monolinoleoylglycerol trimethylsilyl ether	C27H54O4Si2	498	2.3	45.8	–	–	Antifungal, antibacterial and antioxidant47	Alkaloid
2-(Cyclohexylmethylidene) hydrazine-1-carboth ioamide	C8H15N3S	185	2.93	7.46	–	–	–	Cyclohexyl group
2-Bromotetradecanoic acid	C14H27BrO2	306	–	–	9.35	27.68	Antifungal48	Fatty acid esters
2-Myristynoyl pantetheine	C25H44N2O5S	484	4.56	4.52	–	–	–	Organic acids and derivatives
2-Propanone, 1,3-dihydroxy-	C3H6O3	90	23.35	7.76	23.84	7.8	Antifungal49	Ketoses
3-(Prop-2-enoyloxy) pentadecane	C18H34O2	282	4.56	4.52	–	–	Antifungal and anticonvulsive50	Ketone group
3-(Prop-2-enoyloxy) tridecane	C16H30O2	254	4.56	4.52	–	–	Anti-bacterial and anti-biofilm51	Acrylate group
3-tert-Butylsulfanyl-3-fluoro-2-trifluoromethyl-a crylic acid methyl ester	C9H12F4O2S	260	2.79	11.17	–	–	–	Acrylic ester
4-(Prop-2-enoyloxy) tetradecane	C17H32O2	268	4.56	4.52	–	–	–	Acrylate
6-Oxa-bicyclo [3.1.0] hexan-3-one	C5H6O2	98	5.01	8.99	–	–	Antimicrobial and antioxidant52	Cyclic ketones
7-Methyl-Z, Z-8,10-hexadecadien-1-ol acetate	C19H34O2	294	–	–	3.8	4.3	–	Fatty acid acetates
8,11-Octadecadiynoic acid, methyl ester	C19H30O2	290	2	3.16	–	–	Antifungal53	Fatty acid methyl esters
9-(3-Fluoro-phenyl)−12-imino-10,11-dioxa-tricy clo [6.2.2.0(1,6)] dodecane-7,7,8-tricarbonitrile	C19H15FN4O2	350	3.07	35.55	–	–	–	Polycyclic organic compounds
9,10-Secocholesta-5,7,10(19)-triene-1,3-diol, 25-[(trimethylsilyl)oxy]-, (3á,5Z,7E)-	C30H52O3Si	488	5.16	27.69	3.8	4.3	Anticancer54	Steroid family
9,12,15-Octadecatrienoic acid, 2-(acetyloxy)−1-[(acetyloxy) methyl] ethyl ester, (Z, Z,Z)-	C25H40O6	436	–	–	3.8	4.3	Antimicrobial and anti-inflammatory55,56	Fatty acid derivative
9-Octadecenoic acid, (2-phenyl-1,3-dioxolan-4-yl) methyl ester, cis	C28H44O4	444	5.16	27.69	9.35	27.68	Antimicrobial and anti-inflammatory56,57	Fatty acid derivative
à-d-Glucofuranosyl benzenesulfonate	C12H16O8S	320	3.07	35.55	–	–	–	Organic sulfonates
Androstane-11,17-dione, 3-[(trimethylsilyl)oxy]-, 17-[O-(phenylmethyl) oxime], (3à,5à)-	C29H43NO3Si	481	2.3	45.8	–	–	Antimicrobial, nti-ainflammatory57	Modified steroid molecule
Benzaldehyde, 4-nitro-, (6-methyl-1,4-diazatricyclo [4.3.1.1(4,8)] undec-7 -yliden) hydrazine	C17H21N5O2	327	–	–	9.35	27.68	–	–
Carbonic acid, (ethyl)(1,2,4-triazol-1-ylmethyl) diester	C6H9N3O3	171	5.01	8.99	–	–	–	Organic carbonates.
Cyclohexanone	C6H10O	98	5.01	8.99	–	–	Antifungal and antibacterial58,59	Organic compounds
Cyclopentanol, 1-[2-methyl-3-(methylthio)−2-propenyl]-, (Z)-	C10H18OS	186	2.93	7.46	–	–	Antimicrobial60	Cyclopentanol family
Cyclopropaneoctanoic acid, 2-octyl-, methyl ester	C20H38O2	310	2.00	3.16	–	–	Antifungal against Sclerotium bataticola61	Fatty acids
Desulphosinigrin	C10H17NO6S	4279	3.37	28.45	–	–	Antiparasitic62	Glucosinolate
d-Glucosamine	C6H13NO5	179	23.35	7.76	23.84	7.8	Antimicrobial63	Amino sugars
d-Glycero-d-ido-heptose	C7H14O7	210	–	–	10.55	15.86	Antibiotics or immune responses, anti-inflammatory and antiseptic64	Carbohydrates (Monosaccharides)
Dimethylmuconic acid	CH10O4	170	3.07	35.55	–	–	Antimicrobial65	Dicarboxylic acid class
Diphenyl sulfone	C12H10O2S	218	3.07	35.55	–	–	Antibacterial66	Organic compounds
dl-Glyceraldehyde	C3H6O3	90	21.92	5.81	27.28	5.86	–	Monosaccharide
dl-Glyceraldehyde dimer	C6H12O6	180	21.92	5.81	27.28	5.86	–	Organic compound
D-Mannopyranose	C6H12O6	180	10.17	15.85	10.55	15.86	Antibacterial67	Monosaccharides
Estra-1,3,5(10)-trien-17á-ol	C18H24O	256	5.16	27.69	–	–	Antifungal61	Steroid
Formic acid, 2-propenyl ester	C4H6O2	86	2.79	11.17	–	–	–	Ester class of organic compounds
Galacto-heptulose	C7H14O7	210	–	–	10.55	15.86	Antifungal68	C-glycosyl compounds
Glycine, N-[(3à,5á,7à,12à)−24-oxo-3,7,12-tris [(trimethyl ssilyl) oxy]cholan-24-yl]-, methyl ester	C36H69NO6Si3	695	1.98	49.65	5.11	50.24	Antibacterial and antiperspirant69	Steroid
Heptasiloxane, 1,1,3,3,5,5,7,7,9,9,11,11,13,13-tetradecamethyl	C14H44O6Si7	504	1.98	49.65	5.11	50.24	–	Siloxane compound
Hexadecanoic acid, 1a,2,5,5a,6,9,10,10a-octahydro-5,5a-dihydroxy-4 -(hydroxymethyl)−1,1,7,9-tetramethyl-11-oxo-1 H −2,8a-methanocyclopenta[a] cyclopropa[e] cyclod ecen-6-yl ester,	C36H58O6	586	2.3	45.8	–	–	–	Fatty acids
Hexasiloxane, 1,1,3,3,5,5,7,7,9,9,11,11-dodecamethyl	C12H38O5Si6	430	1.98	49.65	5.11	50.24	Antibacterial70	Alkaloids
l-Alanine, N-methoxycarbonyl-, butyl ester	C9H17NO4	203	2.93	7.46	–	–	–	Amino acid derivatives
Methanamine, N-hydroxy-N-methyl-	C2H7NO	61	21.92	5.81	27.28	5.86	Antifungal against Candida albicans and Aspergillus niger71	Hydroxylamine derivative
Methyl nitrite	CH3NO2	61	–	–	27.28	5.86	Antibacterial and antifungal72	Alkyl nitrite
N, N-Dimethyl-O-(1-methyl-butyl)-hydroxylami ne	C7H17NO	131	21.92	5.81	–	–	Antifungal73	Hydroxylamine derivative
o-Acetyl-L-serine	C5H9NO4	147	2.93	7.46	–	–	Anti-inducers55,56	L-alpha-amino acids
Octadecanal, 2-bromo-	C18H35BrO	346	–	–	3.8	4.3	Antibacterial and antifungal74	Bromoalkane
Octasiloxane, 1,1,3,3,5,5,7,7,9,9,11,11,13,13,15,15-hexadeca methyl	C16H50O7Si8	578	1.98	49.65	5.11	50.24	Antimicrobial75	Alkaloid
Oxirane, [(2-propenyloxy) methyl]-	C6H10O2	114	2.79	11.17	–	–	Antifungal76	Epoxide and glycidyl ether
Pentasiloxane, 1,1,3,3,5,5,7,7,9,9-decamethyl-	C10H32O4Si5	356	2.87	50.4	–	–	Antibacterial77	Organosilicon compounds.
Phenol, 2,4-bis(1,1-dimethylethyl)-	C14H22O	206	8.51	26.41	15.6	26.47	Antibacterial, antifungal, anticancer and antioxidant48	Substituted phenols
Propanal, 2,3-dihydroxy-	C3H6O3	90	21.92	5.81	27.28	5.86	Antifungal against F. oxysporum44	Aldehyde and alcohol functional groups
Pyracarbolid	C13H15NO2	217	3.07	35.55	–	–	–	–
Thioacetic acid, S-(3,5-dihydroxypentyl) ester	C7H14O3S	178	5.16	27.69	–	–	–	Thioesters

MW = Molecular weight, RT = Retention time.

Fig. 8.

Chromatograms from GC-MS analysis showing the bioactive metabolites present in the culture filtrates of T. asperellum SEPA11A (A) and T. harzianum SEPA11B (B).

In SEPA11A filtrates, the most abundant compounds were (S)-(+)−2-Amino-3-methyl-1-butanol, 2-Propanone, 1,3-dihydroxy-, and d-Glucosamine (23.35%, retention time 7.76 min), followed by dl-Glyceraldehyde, dl-Glyceraldehyde dimer, Methanamine, N-hydroxy-N-methyl-, and N, N-Dimethyl-O-(1-methyl-butyl)-hydroxylamine (21.92%, retention time 5.81 min). In SEPA11B filtrate, the most abundant metabolites (27.28%, retention time 5.86 min) were dl-Glyceraldehyde, dl-Glyceraldehyde dimer, Methanamine, N-hydroxy-N-methyl-, Methyl nitrite, Methyl nitrite, and Propanal, 2,3-dihydroxy-. These were followed by 2-Propanone, 1,3-dihydroxy-, d-Glucosamine, DL-Arabinose, and Mannosamine (23.84%, retention time 7.8 min) (Table 2).

Interaction between T. asperellum SEPA11A and T. harzianum SEPA11B

A compatibility assay was performed to assess the potential of T. asperellum SEPA11A and T. harzianum SEPA11B for co-formulation as bioactive ingredients. Results revealed there was no antagonistic interaction between the two isolates, suggesting they can be combined in a single formulation. This compatibility indicates potential synergistic effects when both isolates are applied together, offering enhanced biocontrol efficacy compared to individual applications.

Shelf-life assessment of the granular biofungicide formulations (GBF)

The viability of the developed biofungicide formulations was monitored for 14 months under two storage temperatures (5°C and 25 °C). Statistical analysis revealed that the mixed formulation (GPF1 + GPF2) with a water activity (a_w) of 0.11 showed the highest shelf stability, maintaining viability up to 10 months at 5 °C (3.19 × 10⁵ CFU g^-1 and 8 months at 25 °C (3.17 × 10⁵ CFU g^-1, with no significant difference from the initial population (4.71 × 10⁵ CFU g^-1. All formulas demonstrated improved viability under low water activity of 0.11 and cooler storage conditions, emphasizing the importance of humidity and temperature control for long-term shelf-life (Table 3).

Table 3.

Effect of storage duration, water activity (a_w), and temperature on the viability of GBF1, GBF2 and the combined formulation (GPF1 + GPF2).

B		5⁰ C								Mean (A)
	A	0	2	4	6	8	10	12	14
GPF1	Control	41 ± 20c−e	34.67 ± 11.37e−i	26.67 ± 11.55j−p	20.87 ± 2.05p−u	20.23 ± 0.64p−w	14.17 ± 10.21u−A	13.87 ± 10.64v−A	13.33 ± 11.55x−A	23.10 a
	0.11	41 ± 20c−e	36.67 ± 11.55d−h	33.33 ± 11.55f−j	26.90 ± 11.09j−p	23.57 ± 11.15k−q	23.38 ± 1.10zA	20.87 ± 13.94p−u	16.73 ± 21.26 q−z	27.81 b
	0.59	41 ± 20c−e	29.17 ± 2.52i−n	17.77 ± 12.48q−y	13.38 ± 11.51w−A	13.36 ± 11.50w−A	10.36 ± 1.10zA	10.25 ± 0.78zA	13.0 ± 1.16A	18.12 b
GPF2	Control	41 ± 20c−e	46.67 ± 11.55a-c	40.02 ± 0.06c-f	30.31 ± 1.21 h-k	23.05 ± 10.13 m-s	20.37 ± 2.55p−v	13.33 ± 11.55x−A	13.0 ± 11.39y−A	28.47 b
	0.11	41 ± 20c−e	45.93 ± 11.28 a-c	36.67 ± 10.85 j-p	26.6 ± 11.43j-p	22.93 ± 10.16 m-s	22.53 ± 8.95m−t	16.40 ± 10.28s−A	13.9 ± 0.12zA	28.25 b
	0.59	41 ± 20c−e	35.0 ± 2.0e-i	26.87 ± 10.85 j-p	22.33 ± 8.08n-t	20.1 ± 0.35p-x	13.33 ± 11.55 x−A	12.57 ± 8.89y−A	10.03 ± 0.12zA	22.65 b
GPF1 + GPF2	Control	41 ± 20c−e	45.33 ± 2.31a−c	34.33 ± 10.26 e−i	29.33 ± 4.16i−m	22.0 ± 6.93o−t	15.73 ± 3.23t−A	13.80 ± 10.58v−A	10.33 ± 1.15zA	26.48 c
	0.11	41 ± 20c−e	49.97 ± 0.12a	48.43 ± 3.30ab	42.07 ± 7.16b−d	36.60 ± 11.43d−h	30.14 ± 0.46h−l	23.49 ± 11.29k−r	18.35 ± 5.80q−y	36.25 c
	0.59	41 ± 20c−e	37.63 ± 8.55d−g	32.57 ± 8.89g−j	28±0.33 ± 5.77i−o	26.67 ± 11.55 j−p	23.30 ± 11.43l−r	16.67 ± 11.55r−z	10.0 ± 0.00zA	27.02 d
Mean (B)		41a	40.11a	32.96b	0.003c	23.17d	19.26e	15.69f	12.82g
LSD at 5%		A: 2.43, B: 2.29, A×B: 6.88
P value		< 0.0001
	B	25⁰ C								Mean (A)
A		0	2	4	6	8	10	12	14
GPF1	Control	41 ± 20a−c	27.17 ± 11.59g−i	23.33 ± 11.55i−k	17.68 ± 8.11l−o	14.13 ± 11.91o−r	6.51 ± 0.86tu	0.00v	0.00v	16.23 e
	0.11	41 ± 20a−c	27.17 ± 11.59g−i	20.00 ± 0.0k−n	16.75 ± 11.62m−o	16.34 ± 11.02m−q	13.68 ± 11.98o−r	11.35 ± 3.02p−t	60.33 ± 0.31tu	18.98 d
	0.59	41 ± 20a−c	20.50 ± 10.54k−m	14.67 ± 9.45n−r	10.22 ± 0.67r−t	6.26 ± 0.72tu	0.00v	0.00v	0.00v	11.58 f
GPF2	Control	41 ± 20a−c	35.93 ± 10.28c−e	35.67 ± 9.87c−e	26.67 ± 11.55g−j	19.97 ± 0.12k−n	7.83 ± 2.52s−u	0.00v	0.00v	20.88 bc
	0.11	41 ± 20a−c	33.39 ± 11.95d−f	29.97 ± 0.12f−h	22.93 ± 10.16i−l	16.67 ± 11.55m−p	12.94 ± 10.15o−s	12.60 ± 11.11o−s	4.44 ± 1.41u−v	21.74 b
	0.59	41 ± 20a−c	26.19 ± 7.96g−j	31.00 ± 20e−g	19.97 ± 0.12k−n	12.93 ± 11.16o−s	0.00v	0.00v	0.00v	15.42 e
GPF1 + GPF2	Control	41 ± 20a−c	34.30 ± 11.79d−f	31.00 ± 20e−g	25.00 ± 8.72h−k	16.33 ± 11.02m−q	11.00 ± 3.46q−t	0.00v	0.00v	19.83 cd
	0.11	41 ± 20a−c	45.50 ± 9.59a	41.73 ± 6.0ab	36.60 ± 11.43b−d	29.60 ± 1.39f−h	22.57 ± 8.89i−l	16.73 ± 11.32m−o	9.87 ± 1.55r−t	30.45 a
	0.59	41 ± 20a−c	30.07 ± 0.23f−h	22.97 ± 10.28i−l	21.53 ± 5.31j−m	13.33 ± 11.55o−r	0.00v	0.00v	0.00v	16.11 e
Mean (B)		41a	31.08b	26.96c	21.93d	16.17e	8.28f	4.52g	2.26h
LSD at 5%		A: 1.90, B :1.79, A×B: 5.38
P value		< 0.0001

*Colony-forming units (CFU) are expressed per gram of formulation. Data are presented as mean ×10⁴ ± SD. Values followed by different letter(s) within the same column indicate significant differences according to Tukey’s test at P ≤ 0.05 (n = 3).

Discussion

Ensuring healthy seeds is a prerequisite for optimal crop establishment and yield, that is indicating that getting healthy seed means that presence of good seed with good yield production. The present study revealed a high incidence of seed-borne mycoflora in sorghum, including both saprophytic and pathogenic fungi. A total of 22 fungal species belonging to 16 genera were isolated using both AP and DFB techniques. The AP technique yielded higher fungal diversity, particularly for A. niger, A. flavus, A. alternate, Fusarium spp., and Bipolaris spp. that have negative effect on seed germination and seed quality due to the mycotoxins/aflatoxins/ochratoxins they produce, which, if consumed by man, cause carcinoma and paralysis⁴⁴. The results are consistent with Mancini et al.⁷⁸ and Parmar et al.⁷⁹, who reported that the AP method generally provides higher sensitivity for detecting a wide range of fungi due to enhanced nutrient availability and clearer colony development. In contrast, the DFB technique is particularly effective for identifying specific, slow-growing pathogens, such as Fusarium and Drechslera, as it reduces interference from fast-growing saprophytes, although it may increase the risk of seed decay. These findings adequately aligned with the reported results of Khalimi & Anik Leana⁴⁴, who emphasized the critical role of plant quarantine, accurate pathogen detection, and seed health testing to prevent the transmission of seed-borne pathogens. Accordingly, seed health monitoring and fungicidal seed treatments should be integrated into national seed certification programs to mitigate yield losses.

The present study indicated that sorghum seeds are heavily infected with key pathogens, including Fusarium sp., C. lunata, and Bipolaris sp., which may cause seed abortion, seed rot, necrosis, rot, and reduced germination⁸⁰. Raising farmer awareness and implementing field inspections and seed health testing programs are essential for improving seed⁸¹. These results are consistent with previous reports that pathogenic fungi, including Fusarium species, can exist as endophytes within healthy seeds^82,83. The frequency of endophyte isolation can vary depending on the culture media used⁸⁴ and the type of host tissue⁸⁵. Endophytic fungi often employ specialized colonization strategies to evade host defenses, allowing them to persist asymptomatically within internal tissues⁸⁶. Liao et al.⁸⁷ defined fungal endophytes as asymptomatic microbial partners intimately associated with healthy plant tissues, capable of conferring benefits, co-evolving with the host, and modulating their lifestyle according to plant developmental stage and environmental conditions. In this context, many fungi typically considered pathogens can inhabit seeds and other plant tissues without causing visible disease symptoms, explaining their frequent detection in externally healthy-looking sorghum grains. This dual lifestyle underscores the importance of distinguishing latent endophytic colonization from active pathogenic infection when assessing seed-associated fungi. So, it is the importance of adequate plant quarantine, correct diagnosis of symptoms and/or methods of detection and isolation of such dangerous pathogen which could be transferred through seeds. Therefore, most countries have to examine seed samples carefully and/or have to treat seeds with fungicides.

Endophytic fungi are widely recognized as beneficial microorganisms that colonize internal plant tissues without causing disease symptoms, forming an integral component of the plant microbiome. Among them, Trichoderma species are among the most frequently reported endophytic and saprophytic fungi, with strong ecological adaptability, rapid growth, and competitive ability in diverse environmental conditions. In the context of seed transmission ecology, certain endophytes can be vertically transmitted through seeds, enabling early colonization of emerging seedlings and influencing initial plant–microbe interactions^40,88,89. This seed-borne habit allows Trichoderma to establish within host tissues from the earliest stages of development, where it can suppress seed-borne pathogens through mechanisms such as mycoparasitism, antibiosis, and nutrient competition. Collectively, these ecological traits—combining efficient tissue colonization, competitive superiority, and early establishment via seeds—explain why Trichoderma remains one of the most reliable and widely studied genera for biological control in sustainable agriculture^90,91. In this study, T. asperellum SEPA11A and T. harzianum SEPA11B were formulated into a bioagent and evaluated for antagonistic activity against C. lunata, and F. verticillioides. Molecular identification based on ITS region sequencing confirmed the identity of both isolates. ITS sequencing remains a cornerstone for fungal taxonomy due to its ability to resolve inter- and intraspecific variation^92,93.

Trichoderma spp. suppresses pathogens via antibiosis, competition, and mycoparasitism, primarily via secretion of bioactive metabolites such as peptaibols, polyketides, terpenes, gliotoxin, and pyrones^63,94. GC-MS profiling of cultural filtrates revealed that both isolates produce a diverse array of antimicrobial metabolites. The GC-MS supports the chemical evidence of the bio-agents cell-free culture filtrates by characterization the bioactive compounds produced for the bioagent metabolites activities in their growing in the interaction zone for mycoparasitism process among both Trichoderma bioagents and each studied fungal pathogens. The most abundant compound in SEPA11A was (S)-(+)−2-Amino-3-methyl-1-butanol (23.35%), a broad-spectrum antifungal agent reported to disrupt fungal cell membranes⁴⁸. Both isolates also produced 2-Propanone, 1,3-dihydroxy- and D-Glucosamine at high levels, which have well-documented antifungal and antibacterial properties^49,63. SEPA11B uniquely produced 2-Bromotetradecanoic acid (9.35%), an antifungal metabolite effective against soil-borne pathogens⁸³. The detection of 2,4-di-tert-butylphenol in both isolates highlights their antioxidant and antimicrobial potential⁵⁷. Collectively, these findings indicate that Trichoderma spp. deploy a complex metabolite arsenal that synergistically contributes to pathogen suppression. Further analysis revealed several low-abundance metabolites with specialized biological functions. Cyclohexanone (5.01% in SEPA11A) demonstrated pronounced antibacterial and antifungal activity⁹⁵. In SEPA11B, galacto-heptulose (10.55%) was detected as an antifungal metabolite known to inhibit the microbial shikimate pathway⁶⁸. Additionally, steroidal compounds such as estra-1,3,5(10)-trien-17α-ol (5.16% in SEPA11A) and androstane-11,17-dione (2.3% in SEPA11A) were detected, both have been linked to anti-inflammatory and anticancer activities⁹⁶. Collectively, these findings suggest that Trichoderma spp. produce a complex metabolite profile in which major, intermediate, and minor components act synergistically to enhance their biocontrol potential. Comparative profiling of SEPA11A and SEPA11B revealed distinct ecological strategies. SEPA11A synthesized a wider spectrum of antifungal metabolites particularly amino alcohols and fatty acid esters, likely conferring superior activity against soil-borne plant pathogens. In contrast, SEPA11B produced unique brominated fatty acids, indicating a specialized mechanism for niche competition and pathogen suppression⁷².

The antagonistic potential of these isolates was further confirmed by dual culture assays. SEPA11A and SEPA11B exhibited the highest antagonism index (1.0) against F. verticillioides and C. lunata SOR7. After eight days of incubation, microscopic observations revealed that both pathogens were completely overgrown by the Trichoderma isolates, with a marked reduction in pathogen growth and complete absence of sclerotia formation compared to the control. These findings may be associated with mechanisms such as the secretion of cell wall-degrading enzymes by Trichoderma species during mycoparasitic interactions, as reported in previous studies⁵⁷; however, this aspect was not directly investigated in the current work. The observed antagonistic effects of Trichoderma isolates may be associated with the secretion of cell-wall-degrading enzymes, such as chitinases and β−1,3-glucanases, during microparasitic interactions with fungal pathogens, as reported in previous studies^97–99. However, no enzyme activity assays were conducted in the current work, and these mechanisms are suggested based on established literature. Our results showed that SEPA11B actively overgrew both target pathogens, which likely facilitated the formation of boreholes in the host hyphae and subsequent nutrient acquisition. Biocontrol Trichoderma isolates are also known to produce diverse toxic secondary metabolites such as pyrones, koninginins, viridin, gliovirin, gliotoxin, and peptaibols, which contribute to pathogen inhibition⁹¹. The microparasitic mechanism of T. asperellum and T. harzianum observed here follows a well-documented sequence: Initial pathogen recognition, emergence of conidiophores and compact conidia between the pathogen hyphae, subsequent hyphal coiling around the pathogen, and finally, cell wall lysis through the secretion of hydrolytic enzymes, including β-glucanase, chitinase, and protease⁹⁷. For identifying functionally dominant metabolites and understanding their synergistic modes of action will be a focus of future research, including bioassay-guided fractionation and combinatorial testing in vitro. Dual culture assays confirmed the strong antagonistic capacity of SEPA11A and SEPA11B, which completely overgrew F. verticillioides and C. lunata colonies and inhibited sclerotia formation. The SEM is used to gain deeper insights into the antagonistic mechanisms of T. asperellum SEPA11A and T. harzianum SEPA11B against the target fungal pathogens. Electron microscopic examination revealed hyphal coiling, lysis, and disintegration of pathogen hyphae, consistent with mycoparasitic behavior. These results are in agreement with previous reports describing Trichoderma-mediated hyphal deformation, cytoplasmic leakage, and induction of plant defense responses¹⁰⁰.

A successful biocontrol formulation containing fungal propagules should be easy to prepare, maintain its functional properties (viability, germination rate, enzymatic activity) during prolonged storage, and facilitate its application on target organisms^97–99, this may be related to ISR. So, as a priority for future investigation, since we plan to perform transcriptomic and gene expression profiling to identify the precise ISR signaling cascades and defense-related genes induced following colonization by SEPA11A and SEPA11B. Trichoderma species, which are free-living fungi, are known for their ability to promote plant vigor, optimize water use efficiency, and improve the uptake of both macro- and micronutrients. They also enhance photosynthetic performance, increase plant height and stem diameter, improve other agronomic characteristics, and boost overall crop yield^20,98,99,101. The plant growth-promoting effects of Trichoderma are largely attributed to the release of biologically active signaling molecules that diffuse through the soil and act in a manner similar to plant hormones^101,102. The formulated bioagent exhibited extended shelf life under low-temperature (4–5 °C) and moderate water activity (a_w ≈ 0.57) conditions, retaining viability for up to 12 months. These results support the notion that proper formulation and storage conditions are critical for maintaining the efficacy of Trichoderma-based bioproducts⁹⁶. Long shelf-life and faster rate of communalization, and success field usage of Trichoderma formulation also depending on the characteristics of Trichoderma such as, high competency of rhizosphere, highly competition with other soil saprophytes, improving plant growth, easily multiplication of mass, action broad spectrum, excellent control, safety to environment and compatibility between other bioagents¹⁰³. Furthermore, the growing demand for extended shelf-life in biofungicide formulations justifies the selection of semolina as a carrier material in the granular product. Semolina (durum wheat flour) was chosen for two primary functions: First, as a nutrient-rich substrate that supports the survival, viability, and gradual growth of Trichoderma during storage and after field application; and second, as a binding agent whose fine particle size and cohesive properties facilitate stable granule formation when combined with inert materials such as kaolin. In addition, cereal-based carriers—including wheat derivatives—are widely recognized in the literature as effective vehicles for fungal biocontrol agents due to their biodegradability, availability, and capacity to preserve microbial viability. Relevant references confirm that this strategy is consistent with established bioformulation approaches^{96,97,103,104}. The granular biofungicide is designed for soil delivery, not foliar spray. Granules can be applied directly to soil (e.g., planting rows or root zone) or as a seed coating after light moistening. Spraying is unsuitable, as the dough-based granules are not water-dispersible and would settle. The formulation intentionally maintains a solid granular structure for gradual release and localized Trichoderma activity in the rhizosphere^{40,91,96,103,104}.

Conclusion

In this study, two novel endophytic Trichoderma strains—T. asperellum SEPA11A and T. harzianum SEPA11B—were isolated from healthy sorghum seeds and demonstrated strong antagonistic activity against the seed-borne pathogens F. verticillioides and C. lunata, the causal agents of sorghum stalk rot and leaf spot, respectively. Their biocontrol potential was supported by the identification of 55 bioactive metabolites via GC–MS and confirmed by severe ultrastructural damage to pathogen hyphae observed through scanning electron microscopy, while a granular formulation of both strains maintained high viability for up to 14 months. These findings highlight the dual potential of SEPA11A and SEPA11B as eco-friendly biocontrol agents whose seed-borne endophytic origin confers unique advantages such as systemic protection and vertical transmission, positioning them as promising candidates for sustainable biofungicide development pending future greenhouses and field validation, and large-scale production optimization. including monitoring of colonization dynamics and persistence under different environmental conditions. Also, as a future priority, we plan to use transcriptomic and gene expression profiling to identify the specific ISR signaling pathways and defense genes induced by SEPA11A and SEPA11B colonization, and quantify ISR-related responses and better distinguish between these mechanisms.

Materials and methods

Pathogenic fungi

Highly pathogenic isolates of C. lunata (SOR7) and F. verticillioides (SOR2) were obtained from the Seed Pathology Research Department, Plant Pathology Research Institute, Agricultural Research Center, Giza, Egypt. These isolates were subcultured and maintained on potato carrot agar (PCA) slants (Trafalgar Scientific and Midland Scientific, England), and incubated at 25 ± 2 °C for 5 days to maintain viability and sporulation prior to use in this study.

Collection of seed samples

Twenty-two healthy-looking sorghum seed samples (500 g each) were collected during September–October 2023 from six Egyptian governorates (Qena, Asyut, Sohag, Luxor, Dakahlia, and Damietta; latitude 26°012′–31°932′ N, longitude 30°098′–33°039′ E. The samples were collected from an area of 50 m × 50 m in a random zig-zag pattern. Samples were labeled, packed in paper bags, transported to the laboratory, and stored at 4 °C until use.

Isolation of endophytic seed-borne mycobiomes

A total of 22 composite sorghum seed samples were analyzed, which is statistically sufficient. Each sample included 25 seeds per Petri plate with 8 replicates, resulting in 200 seeds examined per sample. Analyses were performed using both agar plate (AP) and deep-freezing blotter (DFB) techniques (ISTA 2007)¹⁰⁵, ensuring robust detection of endophytic seed-borne fungi across all sampled regions.

Healthy-looking sorghum seeds are characterized by uniform size, weight, and color, typically featuring a smooth, undamaged, and clean surface free from debris, cracks, or discoloration were surface-sterilized by sequential immersion in 70% ethanol (2 min), 2% sodium hypochlorite (1 min), and 75% ethanol (30 s), followed by thorough rinsing with sterile distilled water for 10 min and air-drying under laminar airflow¹⁰⁶.

For the AP method, 25 seeds were placed on Petri plates containing 20 mL of 2% water agar (8 plates/sample) and incubated for 7 days at 22 ± 2 °C. For the DFB method, 25 seeds were placed on moistened sterile filter papers in sterilized Petri plates (8 plates/sample), incubated at 22 ± 2 °C for 24 h, frozen at −20 °C overnight, and then transferred to a growth chamber under white fluorescent light (12 h light/dark cycles) at 22 ± 2 °C for 5 days. Emerging fungal colonies were transferred to potato dextrose agar (PDA; 200 g potato, 18 g dextrose, 18 g agar, and 1 l distilled water) for purification using single spore and hyphal tip techniques¹⁰⁷. Pure cultures were examined under a compound microscope (40X magnification) and maintained in a refrigerator (4 °C) for further studies.

Morphological characterization

Isolated endophytic fungi were characterized based on their macroscopic traits (growth rate, colony color) and microscopic features (conidiophore structure, phialide morphology, conidial size, and presence/absence of chlamydospores) following standard mycological keys^108–113. Molecular sequencing was not performed due to the large-scale nature and scope of the diversity assessment.

Dual culture assay for antagonism

Thirty-nine of the endophytic Trichoderma spp. from sorghum seeds were selected and screened for their antagonistic activity against both F. verticillioides (SOR2) and C. lunata (SOR7) using the dual culture technique¹¹⁴. A 5 mm agar disc from each Trichoderma isolate (5-day-old) was placed 1 cm from the edge of a PDA plate (9 cm diameter), with a 5 mm disc of the pathogen (F. verticillioides or C. lunata) placed opposite. Controls consisted of each organism grown alone on PDA. All treatments were performed in triplicate, and plates were incubated at 25 ± 2 °C for six days. The percentage inhibition of pathogen growth was calculated relative to the control.

Antagonistic interaction was also rated on a scale of 1 to 5 after 10 days¹¹⁵. 1 = Trichoderma completely overgrew pathogen; 2 = Trichoderma colonized ≥ 2/3 of plate; 3 = both Trichoderma and the pathogen colonized half of the PDA plate; 4 = pathogen colonized ≥ 2/3 of plate; and 5 = the pathogen overgrew Trichoderma.

Morphological and molecular identification of the tested Trichoderma spp

Among the 39 Trichoderma isolates tested, Trichoderma isolates 3 and 35 that exhibited the strongest antagonistic antagonistic in vitro potential effects against the target pathogens and were subjected to detailed for cultural, morphological and molecular identification as following:

Cultural and morphology characterization

For morphological studies of the Trichoderma isolates 3 and 35, growth rates were determined on PDA at 25 ± 2 °C under dark conditions. Mycelial discs (5 mm diameter) were placed in 90-mm Petri plates containing PDA, with three replicates per isolate. Colony diameters were measured after 5 days of incubation. required for the mycelium to completely cover the plate surface was recorded, along with colony morphological characteristics, including colony appearance, color, and sporulation. Microscopic observations were performed using a light microscope (Olympus CX41, Olympus Corporation, Japan), and images were captured using AxioVision MTB2004 software (Carl Zeiss AG, Germany). Fungal isolates were morphologically identified based on macroscopic characteristics (growth rate and colony color) and microscopic features, including conidiophore structure, length of phialides, conidial size and the presence or absence of chlamydospores. Identification was carried out according to the taxonomic keys of Domsch et al.¹¹⁰, Gams and Bissett¹¹¹, Samuels and Hebber¹¹².

Molecular confirmation

The Trichoderma isolates 3 and 35 that cultural and morphology identified as T. asperellum and T. harzianum were molecularly confirmed, their genomic DNA was extracted using the GeneDirex DNA extraction kit (Cat. No. NA023-0100), following the manufacturer’s protocol. DNA purity and concentration were assessed using a NanoDrop 2000c spectrophotometer (Thermo Scientific, USA). The internal transcribed spacer (ITS) region of each Trichoderma isolate was amplified via polymerase chain reaction (PCR) using primers ITS1: (5’-CTTGGTCATTTAGAGGAAGTAA-3’) and ITS4: (5’-TCCTCCGCTTATTGATATGC-3’)¹¹⁶. PCR was performed in 50 µL reactions containing 25 µL 2X PCR Master Mix (Thermo Scientific, USA), 1 µL of each primer, 2 µL DNA (50 ng), and 22 µL nuclease-free water. Cycling conditions: 95 °C for 5 min, 35 cycles at 95 °C for 30 s, 50 °C for 30 s, 72 °C for 1 min, followed by 72 °C for 10 min¹¹⁷ PCR products were visualized on 1% agarose gel electrophoresis using a 50 bp DNA ladder¹¹⁸. Amplicons were purified using the GeneJET Gel Extraction Kit (Thermo Fisher Scientific, USA) and sequenced via Sanger sequencing (Macrogen, Korea)¹¹⁹. The resulting sequences were analyzed using the ChromasPro software (Technelysium, Australia) and compared against GenBank entries using BLAST (Basic Local Alignment Search Tool; NCBI) (http://www.ncbi.nlm.nih.gov/BLAST/)¹²⁰. A sequence similarity ≥ 98% was used for species-level identification¹²¹. Sequences were deposited in GenBank, and a phylogenetic tree was constructed using MEGA12 software with the maximum likelihood method and 1000 bootstrap replications^122,123.

Scanning electron microscopy (SEM)

To examine the mycoparasitic interactions, mycelial samples were collected from the interaction zones of dual cultures involving each T. asperellum SEPA11A and T. harzianum SEPA11B, and the target pathogens, C. lunata (SOR7) and F. verticillioides (SOR2). These samples represent the interface where Trichoderma actively interacts with the pathogen. The specimens were fixed in 2.5% glutaraldehyde and 2% paraformaldehyde in 0.1 M sodium phosphate buffer (pH 7.4) for 12 h at 4 °C, post-fixed in 2% osmium tetroxide for 90 min, and rinsed three times in 0.1 M sodium phosphate buffer containing 0.1 M sucrose. Samples were dehydrated through a graded ethanol series (80, 90, 96, and 100%) and then coated with gold-palladium⁴⁰. Observations were made using a JEOL JSM-6510 LV SEM at 30 kV (Electron Microscopy Unit, Mansoura University, Egypt).

GC-MS analysis of the bio-agents cell-free cultural filtrates

Trichoderma isolates SEPA11A and SEPA11B that have most effective potential antagonism against fungal pathogens were cultured individually in 50 mL potato dextrose broth (PDB) at 25 ± 2 °C for 14 days under static conditions. After incubation, the cultures were homogenized using a blender and filtered through filter paper to remove mycelia. The resulting Trichoderma culture filtrates were filtered through filter paper and further clarified by centrifugation at 5000 rpm for 20 min, and then sterilized through 0.45 μm Millipore filters to obtain cell-free culture supernatants¹²⁴. Bioactive compounds were extracted from these cultural filtrates using an organic solvent (Ethyl acetate) at a 1:1, v/v ratio by vigorous shaking for 20 min, and the organic phase was collected and concentrated under reduced pressure using a rotary evaporator at 50 °C. Concentrated extracts were derivatized using standard silylation reagents (BSTFA) to improve volatility and thermal stability for GC–MS analysis (Bruker Scion 536 SQ, Scion, Holland)^125,126. The derivatized samples were injected in split mode (1 µL) into the Trace GC-TTSQ mass spectrometer (Thermo Scientific 1310 GC, TSQ 9000, Italy) equipped with a TG-5MS capillary column (30 m x 0.25 mm x 0.25 μm). Helium was used as the carrier gas at 1 mL min⁻¹. Compounds were identified by comparing mass spectra with WILEY 09 and NIST 14 libraries¹²⁷.

Compatibility test between Trichoderma isolates SEPA11A and SEPA11B

To assess compatibility, 5 mm discs of SEPA11A and SEPA11B were placed opposite each other (1 cm from plate edges) on PDA. Each isolate grown alone served as a control. Three replicates were used. The cultures were incubated at 25 ± 2 °C for 9 days in the dark. Radial growth and any inhibition zones were recorded¹²⁸.

Preparation of granular biofungicide (GBF)

Trichoderma isolates SEPA11A and SEPA11B were cultured in PDB for 14 days at 25 ± 2 °C, and spore suspensions were adjusted to 3 × 10⁶ spores mL⁻¹ using a hemocytometer under a light microscope. For the preparation of the granular formulations, 1000 mL of each culture filtrate containing fungal conidia was blended and homogenized with semolina (durum wheat flour) and kaolin as carrier materials. Distilled water was incorporated gradually to form a uniform, workable dough. The dough was sheeted using a hand-operated pasta machine, dried at 45 ± 2 °C, ground, and sieved to obtain granules in the size range of 500–2000 μm. Three GBFs were prepared: GBF1 (SEPA11A), GBF2 (SEPA11B), and GBF1 + GBF2 (combined formulation) following the method described by Shabana et al.¹⁰⁴. The described procedures resulted in a stable granular biofungicide, which is well-suited for subsequent downstream assessment. The intended application method is soil delivery, as opposed to foliar spray. More precisely, the granules can be placed directly into the soil—such as in planting rows or near the root zone. Furthermore, they can also function as a seed coating or dressing: after light moistening to promote attachment to the seed coat, the granules retain their suitability for this use. Their performance will be evaluated in our next phase through downstream greenhouse and open-field studies.

Shelf-life assessment of GBFs

GBFs were stored under two water activity (a_w) levels (0.11, and 0.59) and temperatures (4 and 25 °C). Water activity was maintained with saturated salt solutions (LiCl for a_w = 0.11, Mg(NO₃)₂·6 H₂O for a_w = 0.59)^97,129. Viability was assessed every two months for 14 months by serial dilution plating on PDA and counting colony-forming units (CFU) of fungal propagules. Initial CFU counts were used as reference¹³⁰.

Statistical analyses

Statistical analyses were performed using CoStat software version 6.4 (CoHort Software, Pacific Grove, CA, USA)¹³¹. One-way ANOVA was used to evaluate differences among means, and comparisons were made using Tukey’s HSD test. A split-split plot within stripes design was applied to assess the effects of storage duration, water activity (a_w), and temperature on the viability of GBF1, GBF2, and the combined formulation (GBF1 + GBF2), which their data were expressed as mean ± standard deviation (SD). All statistical tests were conducted at a significance level of P ≤ 0.05. All data in the percentage values were arcsine-transformed and Standard Error bars were included only in Figs. 1 and 2.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

DON

Deoxynivalenol

AP

Agar plate

a_w

Water activity

BCAs

Biocontrol agents

BLAST

Basic local alignment search tool

CFU

Colony-forming units

DFB

Deep-freezing blotter

GBF

Granular biofungicide

GC–MS

Gas chromatography–mass spectrometry

ISTA

International seed testing association

ITS

Internal transcribed spacer

PCA

Potato carrot agar

PCR

Polymerase chain reaction

PDA

Potato dextrose agar

PDB

Potato dextrose broth

SEM

Scanning electron microscopy

SEPA

Seed pathology and examination analysis

Author contributions

Sara Mahmoud Arafa, Khalid Mohamed Ghoneem and Yasser Mohamed Shabana conceptualization, formal analysis, investigation and writing original draft. Sara Mahmoud Arafa, Khalid Mohamed Ghoneem and Allam Arafat Megahed formal analysis, methodology and writing original draft. All authors participate in project administration, resources data curation and manuscript proof editing. All authors have read and agreed to the published version of the manuscript.

Funding

Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB). Not applicable.

Data availability

The datasets generated and analysed during the current study are available in the National Center for Biotechnology Information (NCBI) repository, (https://www.ncbi.nlm.nih.gov/nuccore/LC866760.1) and ( https://www.ncbi.nlm.nih.gov/nuccore/LC866759.1).

Declarations

Competing interests

The authors declare no competing interests.

Ethical approval

was not applicable for this study since it exclusively involved using plant and microflora samples, which typically do not require human or animal ethics review.