Bioactivity and chemical screening of endophytic fungi associated with the seaweed Ulva sp. of the Bay of Bengal, Bangladesh

2.1 Collection, identification and extraction of seaweed

Fresh seaweed samples were collected in December 2020 from the coastline of Saint Martin’s Island, Bangladesh, where Ulva sp. grows abundantly. The identification of the seaweed as Ulva sp. was based on the morphology. A sample specimen has been deposited at the Bangladesh National Herbarium (Accession no. DACB-64245). The air-dried seaweeds were dried at 40 °C for 24 h to reduce moisture content. The dried samples were ground and soaked in 100 % methanol for 7 days at room temperature to extract soluble materials, followed by another 5 days of extraction.

2.2 Isolation of endophytic fungi associated with Ulva sp.

The technique used by Chowdhury et al. (2016) was modified to isolate endophytic fungi from Ulva sp. To remove sand and adhering debris, the seaweed samples were coarsely washed with seawater from the collection site before being transported and processed in the lab within the shortest possible time. The seaweed samples were then subjected to surface sterilization, i.e., immersion of samples in ethanol (EtOH, 70 %), sodium hypochlorite solution (NaOCl, 5 %), and finally in EtOH (70 %) sequentially taking 1–2 min in each solution. To eliminate EtOH residues, all algal samples were then washed three times with sterile water. The selected samples were divided aseptically into small pieces (1–1.5 cm long), which were then put on the culture media supplemented with streptomycin sulphate (100 mg l⁻¹, to suppress bacterial contamination), and then incubated at room temperature in the dark. A few positive controls (media with unsterilized algal samples), a few negative controls (media without any algal samples) and some imprints of sterilized algal samples on the culture media were also incorporated to detect endophytic fungi and to test the effectiveness of the surface sterilization process. The culture medium was prepared by dissolving agar (16 g l⁻¹) in artificial seawater (Nagano et al. 2009; Robinson 1954) and autoclaving at 121 °C for 15 min. The hyphal tips that developed on the initial cultures were transferred to potato dextrose agar (PDA). Pure cultures of the isolated endophytic fungi were then obtained by serial dilution or streaking methods (Fergus 1964).

2.3 Identification of endophytic fungi associated with Ulva sp.

Endophytic fungi were identified taxonomically based on macroscopic and microscopic morphological characterization and molecular identification. The morphological characteristics of each isolate, including the rate of growth, mycelium depth and hyphal orientation, colour, texture, elevation, margin, diameter and form of the colony, as well as spore view, were observed after 3, 6, 9 and 12 days. For molecular identification of fungi, genomic DNA was extracted and the entire Internal Transcribed Spacer region (ITS1, 5.8S, ITS2) was amplified and sequenced (Martin and Rygiewicz 2005). At first, a colony from 4 to 7 days of pure culture of each fungus was selected, scratched with a sterile surgical blade and ground to make powder with liquid nitrogen using a pestle and mortar. Then, DNA was isolated using the Maxwell^® 16 LEV Plant DNA Kit (AS1420, Promega, USA) (Cappuccino and Sherman 1996). DNase-free RNase (30 min at 37 °C) was used to remove RNA contamination from the isolated DNA, and this was then stored at −20 °C for further analysis. Then, the ITS region from the isolated DNA was amplified by PCR (Raja et al. 2017), followed by nucleotide sequencing of the amplicons by the Sanger dideoxy sequencing method.

Raw sequence data was processed using BioEdit 7.2 and then compared to those available in the rRNA/ITS database of the National Center for Biotechnology Information (NCBI) GenBank. The nucleotide sequence of each isolate of the present study was tested for its relatedness with similar sequences using the Basic Local Alignment Search Tool (BLAST). Based on total score, query cover and percent identity, the selected sequences were grouped into distinct clades by constructing the phylogenetic tree. The phylogenetic tree was constructed using the Maximum Likelihood method and the Tamura-Nei model (Tamura and Nei 1993) conducted in MEGA-X software (Kumar et al. 2018). The support of each node was assessed using a bootstrap technique with 1000 iterations, and the tree was scaled with branch lengths denoting the number of substitutions per site. A subculture of each isolate has been deposited at the BCSIR Laboratories, Dhaka, Bangladesh.

2.4 Preparation of fungal crude extracts associated with Ulva sp.

Each endophytic fungal isolate was cultured and incubated at 28 ± 2 °C in the dark. The culture medium was prepared by dissolving PDA (39 g l⁻¹) in artificial seawater and autoclaving at 121 °C for 15 min. After 21–28 days, the media with fungal metabolites were frozen at −20 °C after observing the highest mycelial growth. When thawed, the aqueous part of the fungal culture was separated from the fungal mycelia by filtration and then extracted with chloroform three times using a separating funnel. The fungal mycelia were soaked with ethyl acetate. The EtOAc extract (organic part) was prepared by filtration after 7 days and subsequent solvent evaporation at 5-day intervals using a rotary evaporator at 40 °C (Chowdhury et al. 2017; Khan et al. 2016). The dried fungal crude extracts were kept at 4 °C until further analysis.

2.5 Antimicrobial assay

The antimicrobial activity of the crude extracts was assessed by the disc diffusion method described by Bauer et al. (1966) with some modifications. The crude fungal extracts (2 mg) were dissolved in 200 μl of dichloromethane (DCM), and each disc in the agar plate received 10 μl (i.e., 100 μg disc⁻¹) of the prepared extract solution. Two Gram-positive bacterial strains, Staphylococcus aureus (ATCC 9144), Bacillus megaterium (ATCC 13578) and three Gram-negative bacterial strains, Escherichia coli (ATCC 11303), Salmonella typhi (ATCC 13311), Pseudomonas aeruginosa (ATCC 27833) were applied for the antimicrobial assay as pure cultures obtained from the Institute of Food Science and Technology (IFST), BCSIR Dhaka Laboratories, Bangladesh. Two pure fungal strains, Aspergillus niger (ATCC 1004) and Aspergillus flavus (UCFT 02), were collected from International Centre for Diarrheal Disease Research, Bangladesh (ICDDR,B). The experiment was carried out in triplicate. The mean zone of inhibition (in mm) of each extract was compared with that of two standards, antibacterial agent kanamycin (30 µg disc⁻¹) and antifungal agent ketoconazole (30 µg disc⁻¹). Here, DCM (10 μl disc⁻¹) was used to observe if there was any solvent effect on microorganisms.

2.6 Antioxidant activity

Using a slightly modified version of the approach published by Brand-Williams et al. (1995), the antioxidant capacity of the fungal crude extracts was assessed through 2,2-diphenyl-1-picrylhydrazyl (DPPH) radical scavenging activity. The antioxidant capacity of each extract was compared with that of two standards, the potent antioxidants ascorbic acid (vitamin C) and butylated hydroxyanisole (BHA). The crude extracts (1.6 mg) were dissolved in methanol (MeOH, 400 μl) and serially diluted to obtain 200, 100, 50, 25, 12.50, 6.25, 3.125, 1.56 and 0.78 μg ml⁻¹ concentrations. The extract solutions (2 ml) were mixed with 2 ml of a solution of DPPH in MeOH (20 μg ml⁻¹). Before measuring the absorbance at 517 nm, the mixture was kept in the dark at room temperature for 30 min. Scavenging activity (%) was measured using the equation below. Eventually, inhibitory concentration 50 (IC₅₀ values, μg ml⁻¹), i.e., the concentration of each extract that reduced the DPPH absorbance by 50 % was estimated as a measure of their capacity to scavenge the radical:

2.7 Brine shrimp lethality bioassay

The cytotoxicity of the fungal crude extracts was evaluated on brine shrimp nauplii following the technique reported by Meyer et al. (1982). After dissolving in 200 µl DMSO, the extracts (4 mg) were serially diluted to concentrations of 400, 200, 100, 50, 25, 12.50, 6.25, 3.125, and 1.56 μg ml⁻¹. Each test solution was added to 5 ml of simulated brine water containing 10 shrimp nauplii, and the test tubes were then maintained at room temperature for 24 h. Lethal concentration 50 (LC₅₀ values, μg ml⁻¹), i.e., the concentration of each extract that caused a 50 % lethality of the shrimp nauplii, were compared with that of vincristine sulphate. The brine shrimp lethality bioassay is considered a convenient tool for preliminary toxicity assessment.

2.8 Preliminary chemical screening

The crude extracts of seaweed and its associated endophytic fungi were screened through TLC and NMR spectroscopic analysis for the preliminary visualization of chemical constituents. TLC separations were performed on pre-coated silica gel 60, PF₂₅₄, 0.2 mm aluminium foil (Macherey-Nagel, Germany), at 20 % ethyl acetate in toluene. Spots were detected using UV Lamp (Analytik Jena US, USA) at 254 and 365 nm and then 1 % vanillin-sulphuric acid as the spray reagent followed by heating for 5 min at 110 °C (Chowdhury et al. 2016). NMR spectra were recorded using a Bruker Ascend™ 600 spectrometer at room temperature. The chemical shifts are reported in ppm relative to residual solvent peaks.

2.9 Data and statistical analysis

All the bioactivity tests were repeated three times, and data were documented in triplicate. The IC₅₀ and LC₅₀ values were calculated using logistic regression. To identify significant changes relative to the standards, the data from this study were homoscedasticity-analysed, and then a one-way analysis of variance (ANOVA) was performed. Differences between means were accepted as significant at p < 0.05, and were then subjected to pairwise comparisons of group means using Tukey’s post hoc test. The values are shown as mean ± standard deviation. Calculations and graphs were prepared using Microsoft Excel software.

3.1 Identification of Ulva sp.

The seaweed was identified as Ulva sp. based on morphological characterization (Kipp et al. 2022; Peasura et al. 2015). The macroscopic characteristics of Ulva sp. revealed a vivid grass-green tubular frond and unbranched thalli throughout.

3.2 Endophytic fungi associated with Ulva sp.

The isolation and identification of fungal endophytes from Ulva sp. revealed six different fungi that were most closely related to Chaetomium globosum, Nigrospora magnoliae, Curvularia sp., Curvularia moringae, Aspergillus terreus and Collariella sp. (Table 1).

The isolate UE-1 (Figure 1) was a rapidly growing endophytic fungus. The surface of the colony was cottony and white initially, becoming greyish-olive with age. The reverse of the colony was yellowish to reddish brown. The colony diameter on PDA was approximately 7.8–8.0 cm at 28 ± 2 °C after 6 days in culture. Hyphae were septate with large, oval-shaped, brown-coloured ascomata with wavy filamentous hairs. Asci were stalked and club-shaped, containing ascospores. Ascospores were limoniform, single-celled and yellowish-brown. The colony morphology and microscopic observations suggested the identity of the isolate UE-1 to be a species of Chaetomium (Wang et al. 2016), which was confirmed by the nucleotide BLAST report generated by NCBI database (76.4 % similarity to BLAST best hit with 100 % query coverage). Chaetomium globosum (Accession no. NR 144851.1, connected with the marine environment) was discovered in the phylogenetic tree to be the closest to the isolate UE-1, with a bootstrap support of 92 %. Thus, the isolate UE-1 was recognized as Chaetomium globosum, and its morphology matched that of the relevant species.

Figure 1:

Isolate UE-1 (Chaetomium globosum). (A) Surface of colony, on potato dextrose agar after 6 days culture at 28 °C. (B) Reverse of colony. (C) Ascomata after 28 days culture. (D) Asci. (E) Ascus with ascospores. (F) Ascospores. (G) Phylogenetic tree inferred from internal transcribed spacer sequences using maximum likelihood method.

The endophytic fungus UE-2 was initially white, becoming grey with abundant aerial mycelia (Figure 2). Black areas of conidiation appeared with age. Initially, the reverse of the colony was white and gradually became black. Colonies on PDA reached 7.5–7.8 cm diameter after 6 days at 28 ± 2 °C. Mycelia were superficial and immersed, composed of septate, branched, hyaline and brown hyphae with smooth and thick walls. Conidiophores were short, swelling and tapering at the point of conidium formation. The conidia were dark brown and almost round, slightly flattened. Morphological characteristics indicated that UE-2 was a species of Nigrospora genus (de Silva et al. 2021). The nucleotide BLAST result showed 98.2 % similarity to its best hit with 90 % query coverage. Phylogenetic analysis revealed that this isolate was most closely related to Nigrospora magnoliae (Accession no. NR_172443.1, associated with terrestrial plants), sharing a monophyletic clade with a reasonably good bootstrap support of 85 % (Hillis and Bull 1993). As a result of phylogenetic analysis and morphology, isolate UE-2 was identified as Nigrospora magnoliae when compared to other Nigrospora species.

Figure 2:

Isolate UE-2 (Nigrospora magnoliae). (A) Surface of colony, on potato dextrose agar after 6 days culture at 28 °C. (B) Reverse of colony. (C) Mycelia and conidia after 21 days culture. (D) Conidiogenus cells with conidia. (E) Phylogenetic tree inferred from internal transcribed spacer sequences using maximum likelihood method.

The isolate UE-3 initially appeared as a pinkish colony, growing moderately and gradually turning black; the reverse was the same as the surface (Figure 3A–C). The colony diameter on PDA was approximately 4.5–5.0 cm at 28 ± 2 °C after 6 days of culture. The colony was circular with an entire margin and a woolly texture. Hyphae were long, branched, septate and dark. At the site of conidium development, conidiophores were either simple and straight, branching and curved, or knobby. Conidia were big, typically had four cells, and ultimately took on a curved appearance from the enlargement of a central cell after 5–7 days of ageing. Conidia vary from those of Bipolaris sp. in that they have a darker central cell than the peripheral cells, a finer cell wall, thinner septa between cells, and a distinctive curve developed with maturity. These morphological characteristics indicated UE-3 as Curvularia sp. (Walsh et al. 2018).

Figure 3:

Isolates UE-3 (Curvularia sp., A–C) and UE-4 (Curvularia moringae, D–H). (A) Surface of colony, on potato dextrose agar (PDA) after 6 days culture at 28 °C. (B) Reverse of colony. (C) Mycelia and conidia after 7 days culture. (D) Surface of colony, on PDA after 12 days culture at 28 °C. (E) Reverse of colony. (F) Mycelia and conidia after 7 days culture. (G) Conidium. (H) Phylogenetic tree inferred from internal transcribed spacer sequences using maximum likelihood method.

The surface colour of isolate UE-4 was grey or black with a woolly surface (Figure 3D–F). The reverse of the colony was dark. The growth rate was moderate; the colony diameter on PDA was approximately 2.5–3.5 cm at 28 ± 2 °C after 6 days of culture. Microscopic examination revealed mycelia and conidiophores with conidia, similar to those of isolate UE-3. Moreover, the nucleotide BLAST report generated from the ITS sequence of isolate UE-4 supported this assumption and showed 97.8 % similarity to BLAST best hit with 95 % query coverage. The phylogenetic tree clearly showed that UE-4 was closely related to Curvularia moringae (Accession no. NR_171998.1, associated with terrestrial plant) with a bootstrap value of 88 %, indicating a well supported monophyletic clade. Given these observations, UE-4 was specified as Curvularia moringae among the other genetically similar Curvularia species.

The isolate UE-5 was a filamentous endophytic fungus with a powdery texture (Figure 4). Its growth rate was moderate, reaching 3.0–3.5 cm diameter on PDA at 28 ± 2 °C after 6 days of culture. The surface of the colony was at first white, and then a shade of yellow spreading from the centre and finally turned into cinnamon brown. The reverse of the colony was yellow to brown. Hyphae were septate along with short and smooth conidiophores. Phialides and metulae with chains of conidia were arranged solely on the top half of the vesicle, compact and column-like. Conidia were smooth and rounded. Microscopic and colony morphology indicated that the isolate was a species of Aspergillus (Walsh et al. 2018). The nucleotide BLAST result showed 100 % similarity to its best hit with 96 % query coverage. In the phylogenetic tree, UE-5 had the closest evolutionary relationship with Aspergillus terreus (Accession no. NR_131276.1, associated with indoor dust), forming a monophyletic clade with a bootstrap support of 66 %. Thus, the molecular verification distinguished UE-5 as Aspergillus terreus among other Aspergillus species.

Figure 4:

Isolate UE-5 (Aspergillus terreus). (A) Surface of colony, on potato dextrose agar after 6 days culture at 28 °C. (B) Reverse of colony. (C) Mycelia, conidiophores and conidia after 5 days culture. (D) Conidiophore with conidia. (E) Phylogenetic tree inferred from internal transcribed spacer sequences using maximum likelihood method.

Colonies of the isolate UE-6 on PDA (Figure 5A–C) were translucent to off-white, with an entire or slightly undulate edge, about 3.0–3.5 cm diameter at 28 ± 2 °C after 6 days of culture, and irregularly produced white aerial hyphae and dark ascospores later; the reverse was the same as the surface view. Ascomatal hairs were brown, septate, relatively seta-like or flexuous. Ellipsoidal ascospores were brown. In phylogenetic analysis, UE-6 belonged to a monophyletic clade along with two highly similar species, Collariella gracilis and Collariella virescens, supported by a bootstrap value of 61 % and the best hits in the nucleotide BLAST with 88 % query coverage showed 94.3 % and 94.1 % similarity, respectively, to these species. The two Collariella species shared a 98 % bootstrap support value, as shown in Figure 5D. According to von Arx et al. (1986), these Collariella species are very similar, and differ only in ascomata and ascospores. Due to morphological and molecular similarities, these species were previously known as Chaetomium. Isolate UE-6 was phylogenetically identified as either Collariella gracilis or Collariella virescens (Accession no. NR_147670.1 and NR_147671.1, respectively, associated with soil and rock), whose morphology was also found to be similar to the isolate (Wang et al. 2016).

Figure 5:

Isolate UE-6 (Collariella sp.). (A) Surface of colony, on potato dextrose agar after 12 days culture at 28 °C. (B) Reverse of colony. (C) Terminal ascomatal hairs with ascospores after 45 days culture. (D) Phylogenetic tree inferred from internal transcribed spacer sequences using maximum likelihood method.

3.3 Biological activities of fungal crude extracts

In antimicrobial, antioxidant and brine shrimp lethality bioassays, screening secondary metabolites generated by several marine endophytic fungi under the same fermentation conditions revealed that each fungal crude extract demonstrated one or more pertinent biological activities.

3.3.1 Antimicrobial assay

The study of the fungal crude extracts’ capabilities against pathogenic microbes revealed that most extracts showed some inhibition of single or multiple microbial growth, although kanamycin and ketoconazole exhibited the most potent inhibitory activity against bacteria and fungi, respectively (Figure 6). However, most extracts were more potent against bacterial strains than fungal ones, and UE-2 was the only extract to inhibit the growth of both bacteria and fungi. The solvent (dichloromethane) exhibited no zone of inhibition, as expected. From a broad perspective, the extracts of isolates UE-1, -4, and -6 inhibited the growth of Bacillus megaterium and Escherichia coli more effectively. UE-6 showed the highest antibacterial activity (14.33 mm) against Salmonella typhi growth. UE-1, classified as Chaetomium globosum, provided the only extract that inhibited the growth of Pseudomonas aeruginosa.

Figure 6:

Antimicrobial activity of the crude extracts obtained from marine endophytic fungi associated with Ulva sp. against five bacteria (Staphylococcus aureus, Bacillus megaterium, Escherichia coli, Salmonella typhi, Pseudomonas aeruginosa) and one fungus (Aspergillus flavus). Values are mean ± standard deviation, n = 3. Bars with different letters are significantly different according to Tukey’s post hoc test at p = 0.05. Note: The solvent control (dichloromethane) showed no inhibition (0 mm). The strongest inhibitory effects were observed with the positive controls kanamycin (S1) and ketoconazole (S2).

3.3.2 Antioxidant activity

The DPPH radical scavenging capacity of the fungal crude extracts of marine endophytes derived from Ulva sp. was compared with the potent antioxidants ascorbic acid and BHA, which exhibited IC₅₀ values of 10.67 ± 0.04 and 7.16 ± 0.07 μg ml⁻¹, respectively (Figure 7A). Among all the extracts examined, those of UE-1 and UE-5 also had low IC₅₀ values (21.58 ± 0.06 and 17.80 ± 0.06 μg ml⁻¹, respectively) for scavenging DPPH radical, but the extract of UE-6 had only mild activity as a DPPH radical scavenger at the experimental concentration, while the other extracts had almost no scavenging activity (IC₅₀ > 500 μg ml⁻¹).

Figure 7:

Antioxidant and cytotoxic potential of the crude extracts of fungi associated with Ulva sp. (A) DPPH (2,2-diphenyl-1-picrylhydrazyl) radical scavenging activity, and (B) brine shrimp lethality bioassay. Values are mean ± standard deviation, n = 3. Bars with different letters are significantly different according to Tukey’s post hoc test at p = 0.05.

3.4 Preliminary chemical screening of crude extracts

Crude extracts were subjected to TLC and NMR spectroscopy for preliminary screening of secondary metabolites (Figures S1–S12 and Table S1). Analysis of the TLC spots and the chemical shifts in the spectra (¹H NMR and ¹³C NMR) of the fungal crude extracts revealed the possible presence of diverse secondary metabolites such as sterols, terpenoids, flavonoids, isocoumarins, anthocyanins, anthraquinones and naphthoquinones or their derivatives.

4.1 Endophytic fungi associated with Ulva sp.

Ulva sp., the green seaweed found abundantly on the coastline, is also populated by endophytic fungal taxa, just like other seaweeds. The identification of Ulva sp. was carried out using morphological analysis. The present investigation revealed six different endophytic fungal isolates associated with Ulva sp. The identity of the isolates of Ulva sp. was established by definitive microscopic characteristics followed by macroscopic (phenotypic) observations. The most closely related species was selected more definitively using molecular (genotypic) analysis, such as by comparing the ITS sequence of fungal DNA (Schoch et al. 2012). Table S2 briefly describes the BLAST best hits of the isolates. Many fungi were misidentified previously due to phenotypic variations (Sarwar et al. 2019; Weiß and Göker 2011). Thus, the phylogenetic analysis of DNA sequence data along with morphological characterization has been proved to be an appropriate way to identify fungi correctly (Horiike 2016; Wang et al. 2022). The identification of six isolates derived from Ulva sp. revealed six different fungal species from five genera (Chaetomium, Nigrospora, Curvularia, Aspergillus and Collariella; Table 1) belonging to four classes, namely, Euascomycetes, Sordariomycetes, Dothideomycetes and Eurotiomycetes.

This is the first report of fungal endophytes in Ulva sp. collected from Saint Martin’s Island in the Bay of Bengal, Bangladesh. However, very few reports are available on the diversity of endophytic fungi in green seaweeds. Chaetomium sp., Phomopsis sp., Acremonium sp., Aspergillus niger and Cladosporium sp. were isolated from Ulva lactuca collected from Kovalam (covelong), Chennai (Ahamed and Murugan 2019), whereas Fusarium semitectum, Paecilomyces lilacinus, Aspergillus flavus, Penicillium expansum, P. roqueforti, Rhizopus sp. and Pythium sp. were isolated from the same seaweed species from Johor, Malaysia (Zainee et al. 2021). Other studies have also reported the fungal endophytes Acremonium fuci, Chaetomium globosum, Emericellopsis enteromorphae, E. phycophila, Monodictys putredinis, Parasarocladium alavariense, P. fusiforme and Penicillium sp. from marine green macroalgae (Gonçalves et al. 2019; Singh et al. 2018). The fungal species isolated in this study have been reported previously from a diverse array of macroalgal hosts (Flewelling et al. 2015), except that Nigrospora magnoliae was first isolated from Magnolia plants collected in China and Thailand (de Silva et al. 2021), and this is the second report of an endophytic isolate characterised as N. magnoliae.

4.2 Biological activities of fungal crude extracts

The present work reflects the first evaluation of the crude extracts of six endophytic fungi associated with Ulva sp. from the Bay of Bengal, Bangladesh. Overall biological activities are summarized in Table 2. The extract of isolate UE-1, identified as Chaetomium globosum, was the most promising amongst all the evaluated extracts, demonstrating moderate antioxidant and brine shrimp lethality activity, and also inhibited the growth of all five pathogenic bacterial strains. The extract of isolate UE-2 appeared to have moderate antifungal and weak antibacterial and cytotoxic activities but had no antioxidant effect. The inhibition zones of the crude extracts may appear low in comparison with those of standards (Figure 6), but these zones may be increased if the specific compounds responsible for antimicrobial potential are later identified and analysed. Thus, isolate UE-2, recognized as N. magnoliae, yielded a noteworthy extract and requires further investigation as this species is only recently known to researchers and there is no prior information on its bioactivity. However, antifungal activity of N. magnoliae has been revealed for the first time in the present study. Both extracts of Curvularia sp. (isolates UE-3 and -4) exhibited moderate brine shrimp lethality and mild to moderate inhibition of bacterial growth but showed neither antifungal nor antioxidant activities. Additionally, isolate UE-5, identified as Aspergillus terreus, revealed prospective antioxidant (IC₅₀, 17.80 μg ml⁻¹) and cytotoxic (LC₅₀, 20.20 μg ml⁻¹) activity. Nevertheless, this extract had a weak inhibitory effect on bacterial growth (approx. 8 mm inhibition zone). Lini et al. (2020) reported the antioxidant (IC₅₀, 20.46 μg ml⁻¹), cytotoxic (LC₅₀, 2.85 μg ml⁻¹), and antibacterial (11–12 mm inhibition zone) activity of an ethyl acetate extract of Aspergillus sp. isolated from the green seaweed Caulerpa peltata. Isolate UE-6, identified as Collariella sp., produced an effective extract that might benefit from continued pharmaceutical development, since it exhibited powerful cytotoxicity (LC₅₀, 8.43 μg ml⁻¹), with slight antioxidant capacity and mild to moderate antibacterial effects. The IC₅₀ and LC₅₀ plots are presented in Figures S13 and S14, respectively.

Bioactivity screening of endophytic fungal extracts derived from macroalgal sources opens the door to future possibilities. Antibacterial, antifungal and antioxidant activities have been revealed in several endophytic fungi (e.g., Aspergillus sp., Candida sp., Chaetomium sp., Curvularia sp., Nigrospora sp., Penicillium sp., Pichia sp., Pythium sp. and Rigidoporus sp.) associated with seaweeds (e.g., Caulerpa scalpelliformis, Chaetomorpha sp., Dictyota dichotoma, Euchema sp., Gracilaria edulis, Halymenia sp., Sargassum sp., Stoechospermum marignatum and Ulva sp.) (Ahamed and Murugan 2019; Ravindran and Naveenan 2011; Suryanarayanan et al. 2010; Vega-Portalatino et al. 2023). Teixeira et al. (2019) revealed antitumour compounds in endophytic fungi derived from green macroalgae, such as, Aspergillus versicolor, Chaetomium globosum, Coniothyrium cereal, Gibberella zeae and Penicillium sp.

Even though there has been a lot of bioprospecting research on endophytic fungi associated with seaweeds, we still know very little about the diversity of fungi in marine ecosystems, the bioactive potential of fungal species, and the broader taxonomic groups that can be found in the Bay of Bengal, Bangladesh (Lini et al. 2020). The biological activities screened in this study have not been tested before with endophytic fungi associated with Ulva sp. The crude extracts from these isolates, cultured under identical circumstances, showed some notable variations in the bioactivity. Moreover, based on the variety in the bioactivity of endophytic fungi described here, exploring additional potential bioactivities and assessing how the fungal growth environment affects the generation of its secondary metabolites is crucial.

4.3 Preliminary chemical screening of crude extracts

The evaluation of TLC profiles and NMR spectroscopic data of crude extracts has revealed the existence of diverse metabolites. Visualization of multiple spots on the TLC plate (Figure S1) and ¹H NMR and ¹³C NMR experiments (Table S1, Figures S1–S12) has indicated substance classes such as anthraquinones, naphthoquinones, anthocyanins (Chowdhury et al. 2017; Khan et al. 2018), terpenoids, steroids (Cohen et al. 2011; Ericsson and Ivonne 2009), flavonoids (Sohrab et al. 2004), isocoumarins (Krohn et al. 2004) and their derivatives (Harborne 1998; Liu 2021; Mahmud et al. 2020). The further isolation of prospective metabolites could be guided by the TLC profile or NMR resonance peaks of the crude extracts. Different crude extracts in this experiment developed different patterns on the TLC plate and NMR spectra, which might be promising contributors to bioactive metabolites on further analysis.

Our study focused on the isolation of the endophytic fungal inhabitants of Ulva sp., a marine green macroalga found abundantly in the Bay of Bengal, Bangladesh, and the recognition of the most bioactive fungal species by performing preliminary analyses of their crude extracts. Six distinctive fungal endophytes were isolated from Ulva sp. for the first time in Bangladesh. In addition, the isolates were identified by comparing their ITS sequences with the most closely related species. Most of these isolates (except UE-5) also showed greater or smaller dissimilarities from their BLAST best hits to the nearest species. It would be worth investigating their taxonomy further in case the species isolated have not been sequenced or described before because the taxonomy of marine fungi is still understudied (Calabon et al. 2023; Pang et al. 2023; Pham et al. 2021; Weigand et al. 2019).

Although the seaweed Ulva sp. was identified based on morphology, a molecular analysis to determine the exact species is needed. Moreover, the preliminary analyses suggest that fungal endophytes from Ulva sp. produce secondary metabolites with a wide range of biological properties. The variety and potential of the bioactivities and chemical profiles displayed by the crude extracts of isolates related to Nigrospora magnoliae, Chaetomium globosum and Curvularia moringae have demonstrated the most attractive candidates for further analysis. Subsequent investigations should be conducted to unveil other potential pharmacological activities of these fungal crude extracts of marine origin. Future research should also focus on the determination of bioactive compounds from the host seaweed and its associated fungi.