Column Chromatographies (CCs) of silica gel G (200–300 mesh, Merck.) and Sephadex LH-20 (Merck.) and TLC of silica gel GF254 (10–40 lm, Merck.) were performed. Distillation of all the solvents was done before use. NMR spectra and ESI-MS were obtained using a spectrometer (Bruker AM-400) and Finnigan LCQ-Advantage (m/z), respectively. The chemical shifts d (ppm) rel. to Me4Si, and coupling constants J (Hz.) were then conducted. All the other chemicals under study were of analytical grades.

Samples were collected from five distinct provinces of Iran, representing the versatility of the country, i.e., Esfahan (South of Iran), Mazandaran (North of Iran), Hamedan (West of Iran), Tehran (Center of Iran). Plant temple (stem, leaf) were collected from three species V. major, V. herbacea, and V. minor as listed in Table 1.

Fresh tissue was collected from Apocynaceae, in each locality from June to October, 2014. From each plant specimen, three randomly selected pieces were surface sterilized and in total 700 plant pieces (~0.5 × 0.5 × 1 cm from inner layers) were incubated on potato dextrose agar (PDA) and water agar (WA) culture media, at 26–28°C, for 2–12 weeks. Hyphal tips were isolated, purified, and maintained at 4°C. The endophytic isolates were identified by investigating their colony morphology and the mechanism of spore production on PCA (Potato Carrot Agar), after 7 days at 22–25°C, under 16/8-h light intervals. Fungal specimen was stained and studied under microscope, according to the reference (Bissett, 1984, 1991a,b, 1992; Gams and Bissett, 1998).

The cultures were grown on 2% Malt Extract Agar (MEA) and Potato Dextrose Agar (PDA) at 20°C under an ambient daylight condition or in a light/dark cycle of 12 h/12 h under fluorescent and near-UV light. To determine linear growth rates, fresh mycelial plugs were placed near the edges of PDA plates with a diameter of 9 cm and incubated at 20°C. The colony radius was measured at an interval of 24 h. Using the preparations performed in lactic acid, microscopic observations and measurements were conducted. Following maturing of the conidia within 4–7 days of incubation, observation of the structures and morphologies of macronematous conidiophores taken from the edges of conidiogenous fascicles or pustules was made. After 14 days, conidial morphology was recorded and measurements were done. Based on a comparison with the keys and descriptions presented in recent taxonomic literature, morphologies of the preliminary species were identified through observation (Bissett, 1984, 1991a,b,c, 1992; Gams and Bissett, 1998; Hoyos-Carvajal et al., 2009). Molecular identification was based on the fungal isolate growth in the test tubes containing PDB media at 28°C for 7 days. Using CinnaPure-DNA (Sinaclon, Iran), the genomic DNA was extracted after harvesting the mycelium. A region of nuclear rDNA, containing the internal transcribed spacer regions 1 and 2 and the 5.8S rDNA gene region and a fragment of tef1 was amplified by polymerase chain reaction (PCR) using the primer pair ITS1 (5′-TCCGTAGGTGAACCTGCGG-3′) and ITS4 (5′-TCCTCCGCTTA TTGATATGC-3′) and the primer pair tef1 fw (5′-GTGAGCGTGGTATCACCATCG-3′) and tef1 rev (5′-GCCATCCTTGGAGACCAGC-3′) was done by Kraus et al. (2004). The PCR products were sent to the Macrogen sequencing service (Macrogen Inc. Seoul, Korea) for direct sequencing of double strands of DNA.

Sequences were checked with BioEdit v. 7.0.9.0 (Hall, 2006). The ITS and EF1-α sequences of outgroup (Hypomyces subiculosus TFC 97-166) and additional isolate was retrieved from GenBank. Sequences were aligned with MUSCLE (Edgar, 2004). Manual adjustments were done if necessary after checking the alignments. Using the simple indel-coding implemented by GapCoder, the phylogenetic analyses were performed based on the phylogenetic information contained in the indels (gaps) (Young and Healy, 2003). Phylogenetic analyses were performed with PAUP v. 4.0b10 (Swofford, 2003) for neighbor-joining (NJ) and maximum-parsimony (MP) analyses as described by Abdollahzadeh et al. (2010, 2014). Bootstrap analysis was done with 1,000 replicates. The general time-reversible model of evolution (Rodriguez et al., 1990), including estimation of invariable sites and assuming a discrete gamma distribution with six rate categories (GTR+I+Γ) was used. A partition homogeneity test (PHT) was used to determine the congruence between the ITS and EF1-α datasets (Farris et al., 1995; Huelsenbeck et al., 1996; Abdollahzadeh et al., 2014). New sequences were deposited in GenBank (Table 2).

To cultivate the fungi on PDB, the selected endophyte cultures were inoculated in 250 mL Erlenmeyer flasks containing 100 mL of the medium. Each flask was incubated at 28°C for 2 weeks with periodical shaking at 150 rpm. After the incubation period, the fungal fermentation broth was homogenized by addition of 10% methanol. Methanol and ethyl acetate were used as organic solvents to extract the metabolite via the solvent extraction procedure. An equal volume of solvent was added to the filtrate, mixed well for 10 min and kept for 5 min until the two clear immiscible layers were formed. By using a separating funnel, separation of the upper layer of the solvent, which contained the extracted compounds, was done. To obtain the crude metabolite, the compound yielded by evaporating the solvent was dried in a rotator evaporator under vacuum (Bhardwaj et al., 2015; Sharma et al., 2016). A brown gum was presented after evaporation of the combined ethyl acetate and methanol extracts. The crude extract was then dissolved in dimethyl sulphoxide at 1 mgmL⁻¹of concentration and was stored at 4°C for 24 h before injecting in to GC-MS and the secondary metabolites were kept at −20°C until they were needed for bioassays.

Anti-proliferative and cytotoxic bioactivity of methanol and ethyl acetate extracts was determined against the conidial germination of P. oryzae, as a model. P. oryzae conidial suspension (4 × 104 mL⁻¹; 50 μL including 0.02% yeast extract) was seeded into each well of the 96-well microtiter plates. To obtain the final concentrations of 250, 125, 62.5, 31.25, 15.62, and 7.81 μg mL⁻¹, each well received 50 μL of the sample extract in a serially dilution manner. The assay plates were incubated at 28°C for 16 h. For each sample extract, a microscopic observation was done on the germinations and sizes of the germ tubes that originated from the 75 conidia and the results were compared with the control group to determine MICs. The experiments were performed in triplicate (Kobayashi et al., 1996).

For the measurement of cell viability, tetrazolium salt MTT was applied to determine the fungal methanol and ethyl acetate extracts cytotoxicity (μg/mL⁻¹) against P. oryzae conidia(Levitz and Diamond, 1985; Patel et al., 2013). Our methodology was based on a catalyzed reaction of the functional hyphae through hydrogenases, which led to the cleavage of the yellow tetrazolium salt MTT [3-(4,5-dimethylthiazol-2-yl)- 2,5-diphenyltetrazolium bromide] to MTT-formazan as its purple derivative. MTT-formazan can be quantified through spectrophotometry within 550 nm after being dissolved in isopropanol. A 50 μL of the conidial suspension of P. oryzae (7.5 × 104 mL⁻¹) including a 0.02% yeast extract together with a 200-μl aliquot of conidial suspension was poured into each well of a 96-well plate to yield a final concentration of 15,000 conidia/well. An untreated conidial suspension of fungal extract extracted from P. oryzae was used as the control.

Methanol and ethyl acetate extracts prepared from the endophytes were examined for their antimicrobial activity against Human pathogenic bacteria gram-positive Staphylococcus aureus PTCC (1189) and Gram-negative Escherichia coli PTCC (1399) and Plant pathogenic bacteria Gram-negative Ralstonia solanacearum and Gram-positive Clavibacter michiganensis (laboratory of bacteriology, Bu Ali Sina University). First, bacteria were grown to obtain 1 × 10⁶ CFU mL⁻¹. Then, micro broth dilution assays were performed as described for Anti proliferative Assays, but in nutrient broth (NB) medium. The experimental plates were incubated at 28°C for 16 h for Plant pathogenic bacteria and 37°C for 16 h for Human pathogenic bacteria. The growth of target bacteria was observed and compared with the control to determine the MIC and the minimum bactericidal concentration (MBC). The experiments were performed in triplicate. Upon obtaining the required data, the metabolite concentration required for 80% of in vitro inhibition was presented as IC₈₀ value.

According to the information provided in Tables 3–5 between methanol and ethyl acetate extracts, crude ethyl acetate extract of T. koningiopsis VM115 isolate was used for TLC and GC-MS analysis. The ethyl acetate extract of each isolated fungus was introduced to silica gel plate (10 × 5 cm) by adding a fluorescent indicator, which was developed by placing it in a paper-linked filter in a glass chromatography tank containing 50 ml of an evenly mixed solution of CHCl₃/ MeOH (v/v, 10: 1). As soon as the solvent height was observed to reach 9.5 cm, the plates were removed from the tank. The presence of UV-absorbing compounds was corroborated after seeing dark spots on the dried plates under a UV radiation of 254 nm. The spots were marked with a pencil. Then, using an aerosol spray, a suspension of 10⁶ P. oryzae spores per ml was gently and evenly sprayed onto the plates to make it turn translucent. In a biological hood, the fungal suspension was prepared and sprayed on to the plates using gloves. Washing of the spray bottle with 70% ethanol and sterile distilled water was followed prior to use. The plates were placed in plastic bags in a light box at 28 ± 1°C for a 12-h photoperiod. Within 1 week, the plates were photographed under UV at 254 nm and the inhibition zones were observed as white circular areas of reduced densities or as the fungus non-growth area.

After obtained following the process described above, a culture filtrate of T. koningiopsis VM115 was extracted with ethyl acetate. A residue (20.0 g) was produced in vacuo evaporation of the organic solvent from the extract and dissolved in a lower amount (10 ml) of MeOH around 45°C. The resultant liquor was stored at −10°C overnight, followed by filtration to remove any waxy materials such as the precipitate. MeOH removal under reduced pressure provided a brown lump, and MeOH was added to it until it was completely dissolved. Acetone was added to the MeOH solution drop by drop with acetone; finally at 15% (v/v), it was kept overnight below −10°C to precipitate salts and saccharides. The obtained filtrate was concentrated in vacuo to render a residue (13 g), which was then exposed to column chromatography over Si gel column (100 g, 200–300 mesh) eluting with petroleum ether-acetone mixtures (v/v, 1: 0, 20: 1, 10: 1, 5: 1, 3: 1, 0: 1, 800 ml each) to provide five fractions. Fr.2 (1.9 g)active to the test fungi by TLC bioautography was rechromatographed over Si gel column only (50 g, 200–300 mesh), eluting with CHCl₃-MeOH gradient (v/v, 100:1, 50:1, 20:1, 10:1, 400 ml each). Fr2 was separated further on a Sephadex LH-20 column, eluting with acetone to achieve eight sub fractions Fr2 (1–8). On a silica gel column (20 g), active Fr2 (3) was then purified and eluted with petroleum ether/ethyl acetate (20:1, v/v) to yield 1.045 g of active compound (1). Compound 1 was identified by spectral analyses including MS, ¹H-NMR, ¹³C-NMR (Liu et al., 2006).

Antifungal activities of compound 1 were examined in vitro by testing P. oryzae, Botrytis cinera and Aspergillus fumigatus through the method described in the literature (Barchiesi et al., 2000) based on ketoconazole co-assay as a positive reference. Each compound was isolated in triplicate.

An Agilent technologies 7890A gas chromatograph connected to a 5975 Cinert MSD was used to identify volatile and semi-volatile secondary metabolites from the crude extract of fungi isolate. Also, an HP-5MS fused silica capillary column was applied (Hewlett-Packard, 30 m × 0.25 mm i.d. 0.25 μm film, cross-linked to 5% phenyl methyl siloxane stationary phase). The entire system was checked using Chemstation software (Hewlett-Packard, version A.01.01). Electron impact mass spectra were recorded at 70 eV, and ultra-high pure HAE (99.999%) gas was used as the carrier gas at flow rate of 1 Ml min⁻¹. The injection volume was found to be 1 μL, and all injections were done in a split-less mode. The injector and detector temperature settings were 250 and 280°C, respectively. Column oven temperature was set initially at 50°C for 5 min, then raised to 260°C (ramp: 4°C/min) and held for 5 min. The database of the National Institute of Standards and Technology (NIST) was applied to interpret mass spectrum of GC-MS with more than 62,000 patterns. Using the data obtained from NIST05 (National Institute of Standards and Technology, US), WILEY 8, and FFNSC1.3 (Flavor and Fragrance Natural and Synthetic Compounds) libraries, the existing bioactive compounds in the extracts were identified through a comparison of mass spectra. The molecular weights and structural components of the test materials were determined. Ultimately, most of the compounds of this fungus were identified and verified by comparing the standard GC/MS data with those of the fungal products.

SAS (Ver. 9.1), statistical software was used to compare the means through the test of Least Significant Difference (LSD). The differences between the varied treatments were specified at 5% level (P = 0.05).

In total, seven endophytic Trichoderma fungal isolates were recovered from six different locations from 500 plant specimens from healthy above ground tissue (leaf, stem) of the Vinca major, Vinca herbacea, and Vinca minor (Apocynaceae family) (Table 1). Endophytes were classified according to morphological traits using the key in Gams and Bissett (1998) (Figures 1–4). By combining the 2 unlinked regions of the genes including ITS/ EF1-α, the phylogenetic analyses were conducted. The internal nodes were observed to highly support the phylogenies that resulted from the stable and reproducible ITS/ EF1-α. Production of the trees with similar topologies was represented by the individual congruent datasets since no significant divergence (P = 0.48) was seen through the partition homogeneity test in PAUP 4.0b10 (Swofford, 2003). Therefore, ITS and EF1-α datasets were combined for analysis. The combined ITS and EF1-α sequences for 17 ingroup and 1 outgroup taxa contained 1,574 characters including alignment gaps, of which 334 characters were excluded, 482 were constant, 557 were variable and parsimony-uninformative and 535 were parsimony-informative. A heuristic search of the remaining 535 parsimony-informative characters resulted in a single most parsimonious tree of 426 steps (CI = 0.76, HI = 0.23, RI = 0.86). Seven isolates from four provinces were sequenced for ITS and tef1 and species were identified according to a combination of morphologic and genotypic characters. The identification details of these isolations and their origins are shown in Table 1, in which 4 species are recognized: Trichoderma asperellum, Trichoderma brevicompactum, Trichoderma koningiopsis, Trichoderma longibrachiatum. Results of the phylogenetic analysis based on ITS and EF1-α sequences are shown in Figure 5. Overall, four fungal species were morphologically identified by ITS and EF1-α sequence analyses.

To primarily screen the antitumor activities, P. oryzae fungus was used as a model target (Kobayashi et al., 1996; Dong et al., 2008; Xu et al., 2009). Accordingly, conidial germination and development of germ tube from P. oryzae was adapted for evaluating anti-proliferative activity of methanol and ethyl acetate extracts Trichoderma metabolites. The results presented in Table 3 demonstrate significant bioactivity for all Apocynaceae endophytic Trichoderma isolates. These findings indicate that the ethyl acetate extracts from endophytic Trichoderma isolates showed higher anti-proliferative effects compared to methanol extracts. Notably, the isolate T. koningiopsis VM115 showed the most significant bioactivity among all isolates. T. koningiopsis VM115 showed that conidial germination was completely inhibited at 250–31.2 μg mL⁻¹ and inhibition of germ tube elongation at a concentration of 31.2–7.8 μg mL⁻¹.

The results of the Cell viability of the extracts of Trichoderma isolates are shown in Figure 6. All Trichoderma isolates metabolites exhibited significant activity against P. oryzae conidia after treatment with ethyl acetate extracts and methanol extracts with an IC₅₀ value at a range of 7.8–31.2 μg mL⁻¹ and 15.6–62.5 μg mL⁻¹, respectively. Also ethyl acetate extracts from all endophytic isolates showed higher cytotoxic effects compared to methanol extracts. In total, ethyl acetate extracts of T. koningiopsis VM115 isolate was the most cytotoxic at 7.8 μg mL⁻¹. However, no inhibitions were found in the untreated P. oryzae conidial suspension. The results of cell viability assay and antiproliferative activity indicated that significant cell growth inhibition for all Apocynaceae endophytic Trichoderma isolates. The data elicited strong likelihood of cell growth inhibition and cytotoxic effects by T. koningiopsis VM115 isolate.

All endophytic isolates were screened for antibacterial activity using two human pathogens (S. aureus PTCC and E. coli PTCC) and two plant pathogens (Ral. solanacearum and Cl. michiganensis) as targets. All Trichoderma isolate metabolites showed bacteriostatic and bactericidal activity against the two Gram-negative and two Gram-positive bacteria targets (Table 4). Ethyl acetate extracts were bacteriostatic with an IC₈₀ value at a range of 7.8–15.6 μg mL⁻¹ and bactericidal with an IC₈₀ value at a range of 15.6–62.5 μg mL⁻¹. The methanol extracts were bacteriostatic with an IC₈₀ value at a range of 7.8–62.5 μg mL⁻¹ and bactericidal with an IC₈₀ value at a range of 31.2–62.5 μg mL⁻¹. Also, T. koningiopsis VM115 and T. longibrachiatum VM 99 were more influential against Gram-negative and Gram-positive bacteria. Ethyl acetate extract had higher-level activity than other extracts, compared with bioactivity between methanol and ethyl acetate extract.

According to the information provided in Tables 3–5 between ethyl acetate and methanol extracts, crude ethyl acetate extract of T. koningiopsis VM115 isolate was used for TLC and GC-MS analysis. The one known antifungal metabolite purified from the ethyl acetate extract T. koningiopsis VM115 was identified as compound 1 (Godtfredsen and Vangedal, 1964) via spectral analyses, such as MS, ¹H-NMR, ¹³C-NMR (Figure 7).

Trichodermin (4b-acetoxy-12, 13-epoxytrichothec-9-ene) = colorless oil. ¹H NMR (CDCl3, 400 MHz) δ: 5.48 (1H, dd, J = 7.8, 3.6 Hz, H-4), 5.31 (1H, d, J = 5.4 Hz, H-10), 3.71 (1H, d, J = 5.2 Hz, H-2), 3.50 (1H, d, J = 5.4 Hz, H-11), 3.02 (1H, d, J = 4.0 Hz, H-13a), 2.73 (2H, m, H-13b, H-3a), 2.46 (1H, dd, J = 15.4, 7.8 Hz, H-3b), 1.81–1.97 (3H, m, H-7a, H-8a, H-8b), 1.62 (3H, s, H-16), 1.33 (1H, brd, J = 12.1 Hz, H-7b), 0.83 (3H, s, H-15), 0.61 (3H, s, H-14). ¹³C-NMR (CDCl3,100 MHz) δ: 170.6 (s, C-1'), 139.8 (s, C-9), 118.3 (d, C-10), 78.8 (d, C-2), 74.7 (d, C-4), 70.1 (d, C-11), 65.2 (s, C-12), 48.6 (s, C-5), 47.5 (t, C-13), 40.1 (s, C-6), 36.3 (t, C-3), 27.7 (t, C-8), 24.1 (t, C-7), 23.0 (q, C-16), 20.9 (q, C-2'), 15.7 (q, C-15), 5.5 (q, C-14).

ESI-MS: 315 [M + Na]⁺, 607 [2 M + Na]⁺, 251 (M+H)⁺.

T. koningiopsis VM115 compound was bio assayed against pathogenic and model fungi B. cinera, A. fumigatus, and P. oryzae by the serial dilution method (Table 5). Having an MIC close to that of ketoconazole as a positive control, compound 1 was seen to be involved in a strong bioactivity against the test fungi.

Crude ethyl acetate extract of T. koningiopsis VM115 isolate was used for GC-MS analysis. The mass spectra of GC-MS-MS were interpreted by the database of National Institute of Standards and Technology (NIST) with more 62,000 patterns. The chromatogram anticipated the presence of many compounds and they were identified according to peak area, the retention time, and molecular formula. The retention time and abundance of the compounds under the described conditions in GC-MS section are shown in Table 6; the identification of these metabolites revealed that endophytic T. koningiopsis VM115 has the capacity to produce bioactive compounds. Collectively, each of the seven classes of volatile compounds was generated by T. koningiopsis VM115 [alcohols, esters, monoterpene, pyrones (lactones), acids, furanes, and lipids] (Table 6).The most abundant compound was 6-pentyl-alpha-pyrone (6-PP) with a 20.442 min retention time, based on the total area of the GC analysis, originally specified by Collins and Halim (1972), and indicated as one of the important bioactive compounds from T. harzianum and T. koningii species as reviewed by Hanson (2005).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.