Biodiversity of Aspergillus species in some dried fruits and the traditional food Terkhêna, with description of Aspergillus mahabadiensis sp. nov. from Mahabad, western Iran

INTRODUCTION

Dried fruits, as nutrient-rich healthful snack foods, have been popular in Iran. Historically, they constitute the major part of the Mediterranean diet dates back to the millennial generations in Mesopotamia (incl. western Iran) northern part of the Fertile Crescent. Considering their roles in human health due to their bioactive compounds, micronutrients, fiber content and antioxidant properties daily intake of dried fruits is recommended by the World Health Organization (Alasalvar et al. 2023). In western parts of Iran, raisins and dried fruits such as apple, apricot and white mulberry are the most common traditional snacks. In this part of the country, a special type of traditional dried food known as ‘Terkhêna’ is also very popular cold season food which is made of bulgur wheat, turnip (Brassica rapa L.) or pennyroyal (Mentha pulegium L.) leaves and buttermilk (Mohammadi and Ostovar 2023). Terkhêna is stored dry, and commonly consumed to prevent or treat cold, rhinitis and similar diseases. Since these foodstuffs are processed traditionally and stored in the warehouse for a long time, fungal spoilage and hazards related to their mycotoxins are major concerns for human health (Bhat et al. 2010, Wu et al. 2018, González-Curbelo et al. 2023).

Among the various food spoilage fungi, Aspergillus members are ubiquitous and well-known mycotoxigenic species. The genus Aspergillus, introduced by Micheli (1729), is a diverse, widespread, and economically significant fungus, which seriously affects food and human health. Aspergillus species occur in diverse environments and are found on various substrates at a wide temperature range, as well as under low humidity conditions (Bhat et al. 2010, Krijgsheld et al. 2013). Members of Aspergillus are considered by researchers in various fields of biotechnology and industrial mycology (metabolites production, medicines, fermentation and food processing), food and indoor mycology (food spoilage and mycotoxins production) and medical mycology (human and animal fungal diseases) (Samson et al. 2014).

Taxonomy of Aspergillus has traditionally relied on morphological features, but in recent decades, affected by chemotaxonomic and phylogenetic species concepts and following a polyphasic systematic approach, researchers have used extrolite profiles and DNA sequence data together with phenotypic characters (Samson et al. 2014). The first monographs on Aspergillus were published by Thom and Raper (1945) and Raper and Fennell (1965). Following Raper and Fennell (1965) who divided the genus Aspergillus into 18 groups, infrageneric classification in Aspergillus based on phenotypic data was further developed by Gams et al. (1985) and subgenera and sections were introduced under the genus level, which is supported by current phylogenetic analyses. The taxonomic history has been discussed by Samson et al. (2014) and Houbraken et al. (2020) which updated the taxonomy of Aspergillus in a polyphasic approach. Recent multigene phylogenetic studies often supported by phenotypic, physiologic and/or extrolite data, have listed and classified 453 Aspergillus species in six subgenera, 28 sections, and 76 series (Houbraken et al. 2020, Visagie et al. 2024). Among these, Circumdati with 11 sections (e.g. Flavi, Nigri and Terrei) is the second species-rich subgenus after Nidulantes (Houbraken et al. 2020, Visagie et al. 2024).

Members of section Flavi are of special interest to researchers in terms of food safety, due to the production of some hazardous mycotoxins such as aflatoxins and ochratoxins (Frisvad et al. 2019). They can exist in the form of sclerotia, conidia and mycelium, and can be found in soil, indoor environments and a wide range of foods (Varga et al. 2011). Section Flavi, containing 35 species, resided in subgen. Circumdati and classified into eight series including ser. Flavi. In terms of food safety, Aspergillus flavus/A. oryzae complex is of great importance in ser. Flavi (Houbraken et al. 2020).

Aspergillus species in section Nigri are commonly regarded as safe, and considered in biotechnology and industrial mycology for producing fungal natural products such as enzymes and organic acids (Andersen et al. 2011, Yang et al. 2017) and food fermentation (Hong et al. 2014). However, some species in this section cause food spoilage (Samson et al. 2019) or produce mycotoxins on a wide variety of foods and feeds (Frisvad et al. 2018), and some members are known as minor pathogens that can cause various mycoses in animal and human (Nargesi et al. 2022). In a phylogenetic revision on Aspergillus and allied genera, section Nigri with 29 species placed in subgen. Circumdati and divided into five series including Carbonarii, Heteromorphi, Homomorphi, Japonici and Nigri (Houbraken et al. 2020). Recently, Bian et al. (2022) in an extensive phylogenomic study reduced the number of species in series Nigri from 10 species recognized by Houbraken et al. (2020) to six species, i.e. A. brasiliensis, A. eucalypticola, A. luchuensis, A. niger, A. tubingensis and A. vadensis.

Aspergillus section Terrei was established based on A. terreus, the most well-known and widespread species with great significance in fermentation, food and medicine industries due to the ability to produce a wide range of secondary metabolites such as enzymes, acids, mycotoxins and some drugs, antitumor and antiviral compounds (Samson et al. 2011). Section Terrei species are classified into three series Terrei, Nivei and Ambigui (Houbraken et al. 2020). Series Terrei and Nivei are distinguished based on conidia en masse which are in shades of brown in Terrei and in shades of yellow, vinaceous fawn or white in Nivei. The species in series Ambigui grow slower than the members of series Terrei and Nivei (Houbraken et al., 2020). Thus far, 22 species have been accepted in this section, with most of them being isolated from soil, clinical samples, and a limited number of plant products such as wheat flour, barley, peanuts and corn (Samson et al. 2011, Barros Correia et al. 2020, Houbraken et al. 2020, Wang and Zhuang 2022, Cañete-Gibas et al. 2023). Of these, A. terreus is one of the well-known species that causes invasive aspergillosis with higher mortality rates (Lass-Flörl et al. 2021).

During an extensive study on taxonomy and phylogeny of mycotoxigenic fungi associated with raisins, three dried fruits (apple, apricot, white mulberry) and a traditional food known as ‘Terkhêna’ in Mahabad, located in western Iran, we studied a large collection of 415 fungal isolates belonging to four genera Alternaria, Aspergillus, Penicillium and Rhizopus. Of these, Aspergillus with 317 isolates (76.4%) was the most prevalent genus followed by Penicillium with 50 isolates (12%) belonging to five species (Ghaderi and Abdollahzadeh 2024a), Rhizopus with 33 isolates (8%) identified as R. arrhizus (Ghaderi and Abdollahzadeh 2024b), and Alternaria with 15 isolates (3.6%). In this paper, we focus on the phylogeny and taxonomy of Aspergillus species and introduce a novel species in section Terrei, named here Aspergillus mahabadiensis sp. nov., isolated from samples of dried apricot and the traditional food Terkhêna.

MATERIALS AND METHODS

Sampling and Isolation

During 2021–2022, samples of raisins, dried fruits including apple, apricot and white mulberry, and the traditional food Terkhêna were collected from various locations in Mahabad, Iran. Fungi were isolated by direct plating of surface sterilized (3 min in 70% ethanol) small pieces of the samples on potato dextrose agar (PDA). Pure fungal isolates were obtained using the hyphal tip or single spore techniques on Water Agar (WA 1.5%) and stored on PDA at 4–8 °C. Representative isolates were deposited in the culture collection of the Iranian Research Institute of Plant Protection (IRAN, Tehran, Iran) and the CBS collection of the Westerdijk Institute, Utrecht, The Netherlands.

DNA extraction, PCR amplification and sequencing

Fungal isolates were grown in potato dextrose broth (PDB) for 7–12 days at 20–25 °C. The modified Raeder and Broda (1985) method was followed for DNA extraction from dried mycelium as described by Abdollahzadeh et al. (2009). To manage the costs of DNA sequencing and time for morphological studies, we used the inter-simple sequence repeat (ISSR) technique to cluster fungal isolates based on DNA fingerprinting patterns generated with (GTG)₅ primer (Meyer et al. 1991, Alves et al. 2007). PCR and electrophoresis conditions were adjusted following Alves et al. (2007). DNA fingerprinting profiles were visually analyzed, and a representative isolate from each cluster was selected for further studies. A part of the calmodulin (CaM) gene of representative isolates was amplified and sequenced using primer pairs CF1/CF4 (Peterson et al. 2005). For comprehensive phylogenetic analysis of the potential new species, ITS region (ITS1-5.8S-ITS2, using primer pairs ITS5/ITS4 (White et al. 1990), a part of beta-tubulin (BenA, using primer pairs Bt2a/Bt2b (Glass and Donaldson 1995) and RNA polymerase II second largest subunit (RPB2, using primer pairs 5F2 (Sung et al. 2007) or 5F/7cR (Liu et al. 1999) were amplified and sequenced. The PCR conditions were as follows: an initial denaturation step of 5 min at 95°C followed by 35 cycles of 30 s at 95°C, 30 s at 55°C (ITS, CaM and BenA)/52°C (RPB2), 60 s at 72°C, with a final extension of 7 min at 72°C. Generated sequences in this survey were submitted to GenBank (Table 1).

Phylogenetic analyses

The generated sequences of CaM locus were subjected to BLAST analysis in GenBank. Based on the BLAST analysis, our sequences supplemented with the sequences of the type specimens or authentic specimens of existing species and aligned using online MAFFT v. 7 (Katoh et al. 2019) and edited manually in BioEdit v. 7.0.5 (Hall, 2004), if needed. The same procedure was followed for BenA and RPB2 sequences of the new species located in section Terrei. In section Terrei, both single and multigene phylogenies were conducted. Thus, for multigene phylogenetic analyses, single-locus alignments were concatenated with Mesquite 3.70 (Maddison and Maddison 2023). All single and combined alignments were analyzed by Maximum Likelihood (ML) and Bayesian Inference (BI). In BI analyses, the general time reversible (GTR + G + I) model was used for all single and multigene phylogenies. The BI was executed through the online CIPRES Science Gateway (Miller et al. 2012) using MrBayes v. 3.2.7a (Ronquist et al. 2012), as described by Bashiri et al. (2022). The ML analysis was executed in IQ-TREE v. 2.2.2.7 (Nguyen et al. 2015) For ML analyses the most suitable model was determined for each locus using Modelfinder (Kalyaanamoorthy et al. 2017) and ultrafast bootstrapping executed with 1000 replications using UFBoot (Hoang et al. 2018), both integrated into IQ-TREE. The phylogenetic trees obtained from the analyses were visualized in FigTree v. 1.4.3 (http://tree.bio.ed.ac.uk/software/figtree) and edited in Adobe Illustrator 2023 v. 27.1.0.189. Alignments and trees were deposited in TreeBASE (www.treebase.org, S31425) and taxonomic novelties in MycoBank (www. MycoBank.org, Crous et al. 2004).

Macro and micromorphological observations

The macromorphological characteristics were observed on three point inoculated agar media malt extract agar (MEA), Czapek yeast extract autolysate agar (CYA), yeast extract sucrose agar (YES), oatmeal agar (OA) and Czapek's agar (CZ). Culture media were prepared as described by Samson et al. (2014). All inoculated media were incubated at 25 °C for 7 d, with additional CYA plates incubated at 27 °C, 30 °C, 33 °C and 37 °C. Culture characteristics including colony diameter, texture, degree of sporulation, obverse and reverse colony color (according to Rayner, 1970), production of soluble pigments, sclerotia or ascomata and exudates were recorded after 7 d of incubation. For microscopic observations, mounts were prepared in lactic acid (60%) and excess conidia were removed using ethanol (70%). The shape, color and dimension of the fungal structure were documented using an Olympus BX51 microscope equipped with an Olympus DP72 camera and a measurement module Cell Sens Entry v. 2.1. Dimensions of fungal structures were estimated based on at least 30 microscopic measurements and presented as the range of measurements with extremes in brackets followed by mean ± standard deviation. The recorded images were used to prepare photoplates in Adobe Photoshop 2021 v. 22.5.8.

RESULTS

Sampling, fungal isolation and DNA sequencing

In an extensive phylogenetic study on mycotoxigenic fungi associated with some foodstuffs in Mahabad, western Iran, we obtained 415 fungal isolates from 519 samples of raisins (n = 119), dried fruits including apple (n = 100), apricot (n = 100) and white mulberry (n = 100), and a traditional food ‘Terkhêna’ (n = 100). A total of 317 out of 415 isolates (76.4%) belonged to Aspergillus, including 81 from raisins, 64 from dried apple, 64 from dried apricot, 52 from dried white mulberry and 56 from Terkhêna. Based on DNA fingerprinting profiles generated by (GTG)₅ primer, all 317 Aspergillus isolates were placed in 11 clusters (Fig. 1), and thus 11 representative isolates were selected for phylogenetic analyses.

Table 1. Aspergillus isolates sequenced in this study.

¹ IRAN Iranian Fungal Culture Collection, Iranian Research Institute of Plant Protection, Tehran, Iran; CJA Personal cultures of Jafar Abdollahzadeh. DTO internal culture collection of the Department of Applied and Industrial Mycology of the Westerdijk Fungal Biodiversity Institute, Utrecht, The Netherlands. ^T Ex-type. ² CaM partial calmodulin gene; BenA partial β-tubulin gene; RPB2 partial RNA polymerase II second largest subunit gene; N.S. Not Sequenced.

Fig. 1. DNA fingerprinting patterns of sequenced isolates generated by primer (GTG)₅. Lanes 1-6: A. mahabadiensis sp. nov.; 7: A. falvus (IRAN 5119C); 8: A. terreus (IRAN 5116C); 9-11: A. niger (CJA OGhK108, CJA OGhS9, IRAN 5126C); 13: A. luchuensis (IRAN 5120C); 12, 14-16: A. tubengesis (CJA OGhK4, CJA OGhK106, IRAN 5122C, CJA OGhS40). M: GeneRuler DNA Ladder Mix (100bp).

Phylogeny

BLAST search of CaM sequences revealed that our isolates are placed in sections Flavi, Nigri and Terrei. Based on phylogenetic analyses of CaM sequence data our isolates placed in A. flavus/A. oryzae in section Flavi (Fig. 2), A. luchuensis, A. niger and A. tubingensis in section Nigri (Fig. 3), and A. terreus together with a potential new species in section Terrei (Fig. 4). Identified species with their distribution and number of isolates obtained from each substrate are shown in Fig. 5.

Aspergillus oryzae is known as non-aflatoxigenic domesticated form of A. falvus even though some A. flavus strains have lost their aflatoxin production ability (Yu et al. 2004).  Based on phylogenetic analyses of BenA, CaM and RPB2 sequence data, the two species A. flavus and A. oryzae are conspecific, but due to the importance of aflatoxins production in food safety researchers still know A. oryzae as a legitimate species and differentiate these two species based on the aflatoxin production ability, origin of the strains and morphology (Frisvad et al. 2019). Most of the researchers have identified isolates from fermented foods as A. oryzae and placed isolates from other substrates even with no aflatoxin production ability in A. flavus. In this study, the toxigenic properties of the isolates were not considered and based on origin and morphology some isolates showed an intermediate status. For example, IRAN 5119C morphologically belongs to A. flavus by smaller and rough conidia (3.2–4 µm vs. 4.5–8 µm, rough vs. smooth to rough), degree of sporulation (strong vs. weak) and color of conidia en masse (yellow to green vs. green to brown), while as an isolate obtained from Terkhêna, a fermented traditional food, is an A. oryzae isolate. Thus, with no emphasis on a particular species our isolates are considered as members of A. flavus/A. oryzae complex.

Fig. 2. Maximum likelihood phylogram of Aspergillus section Flavi (series Flavi) based on CaM sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus caelatus NRRL 25528. Strain sequenced in this study is in boldface. ^T = Ex-type strain.

Fig. 3. Maximum likelihood phylogram of Aspergillus section Nigri (series Nigri and Carbonarii) based on CaM sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus flavipes NRRL 302. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Fig. 4. Maximum likelihood phylogram of Aspergillus section Terrei (series Nivei and Terrei) based on CaM sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus ambiguus CBS 117.58. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Fig. 5. Substrates examined: left to right apple, apricot, raisins, Terkhêna and white mulberry (A), identified Aspergillus species (B), and their distribution on each substrate (C–G).

To confirm the potential new species based on representative isolate IRAN 4983C in section Terrei, all other five isolates with the same DNA fingerprinting profile were selected for sequencing in a part of CaM and more DNA barcodes (ITS, BenA, RPB2) necessary for Aspergillus phylogeny. In ML analyses, the nucleotide substitution model TPM2+F+I was used for ITS, K2P + I for BenA, TNe + I for CaM and TIM2e + G4 for both RPB2 and the combined phylogeny. In BI analyses, the general time reversible (GTR + G + I) model was used for all single and multigene phylogenies.

Since ITS sequence data is insufficient for discrimination of all Aspergillus species, both ML and BI analyses resulted in an unresolved tree and our isolates were not correctly identified based on ITS phylogeny (Fig. 6). In both single-gene phylogenies CaM (Fig. 4) and BenA (Fig. 7) our isolates together with seven isolates previously identified as A. cf. allahabadii and A. allahabadii NRRL 4101 (Visagie and Houbraken 2020, Visagie et al. 2024) were placed in a large clade (A) representing a potential new species separate and distant from A. allahabadii. This clade is highly supported in ML (BenA: 94%; CaM: 84%), but posterior probability is lacking in BI analyses (BenA: 0.8; CaM: 0.84). In CaM phylogeny, more resolution was observed and two strongly supported subclades (A1: BS = 96%, PP = 0.99 and A2: BS = 97%, PP = 1) were recognized within clade A in both ML and BI analyses. In RPB2 phylogeny, A1 divided to three separate clades (A1-1, A1-2, A1-3) representing three potential new species, and A2 constituted a distinct clade representing a new potential species as a sister group with CBS 164.63, ex-type strain of A. allahabadii (Fig. 8). Some incongruences were observed between single-gene phylogenies regarding discrimination of the two series Terrei and Nivei (Fig. 4), position of A. barbosae (Figs 7, 8) and the degree of resolution in clades containing A. careus, A. niveus, A. cf. allahabadii isolates from South Africa and our isolates from Iran (Figs 4, 7, 8). However, our isolates together with A. cf. allahabadii isolates were obviously distinguished from the CBS 164.63 ex-type strain of A. allahabadii in all three single-gene phylogenies. Thus, we combined all three alignments and provided a multigene dataset (CaM: 50, BenA: 52, RPB2: 37 sequences). The concatenated alignment consisting of 2095 characters (CaM: 558, BenA: 527, RPB2: 1010), including alignment gaps, was analyzed by Maximum Likelihood (ML) and Bayesian Inference (BI). The ML analysis of the dataset with 733 distinct alignment patterns and 456 parsimony-informative characters, yielded a best-scoring ML tree (lnL = -9953.326). The BI phylogenetic tree was mapped on the ML tree presented in Fig. 9 with ML/BI bootstrap support and posterior probability values at the nodes.

Multigene phylogenetic analyses resulted in a fully resolved tree in both ML and BI analyses. All isolates of A. allahabadii, A. cf. allahabadii and our isolates were placed in a large unsupported clade containing A. allahabadii and four potential new species (A1-1, A1-2, A1-3, A2). In accordance with RPB2 phylogeny, subclade A1 was split in three clades (A1-1, A1-2, A1-3) representing three new species and subclade A2 was placed in a clade as sister group with A. allahabadii. The degree of nucleotide variation in all three loci within and between four potential new species and A. allahabadii is summarized in Table 2. In all single and multigene phylogenies, our isolates were grouped close to isolates CBS 142986 and CBS 142987 and appeared as a candidate for a new species in a subclade with poor (Fig. 4) or strong (Fig. 7) statistical supports or in a well-supported distinct clade (Figs 8, 9). The small degree of nucleotide variation (BenA: one substitution; CaM: four substitutions; RPB2: one substitution) in this clade was interpreted as infraspecies variation. Thus, we recognized this clade as a new species distinguished from all described species in series Nivei and named as A. mahabadiensis sp. nov., which is described here.

Fig. 6. Maximum likelihood phylogram of Aspergillus section Terrei (series Nivei and Terrei) based on ITS sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus ambiguus CBS 117.58. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Fig. 7. Maximum likelihood phylogram of Aspergillus section Terrei (series Nivei and Terrei) based on BenA sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus ambiguus CBS 117.58. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Fig. 8. Maximum likelihood phylogram of Aspergillus section Terrei (series Nivei and Terrei) based on RPB2 sequence data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus ambiguus CBS 117.58. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Fig. 9. Maximum likelihood phylogram of Aspergillus section Terrei (series Nivei and Terrei) based on concatenated dataset of BenA, CaM and RPB2 sequences data. ML/BI ultrafast bootstrap support (≥80 %) and posterior probability values (≥0.95) are shown at the nodes. The phylogenetic tree was rooted with Aspergillus ambiguus CBS 117.58. Strains sequenced in this study are in boldface. ^T = Ex-type strain.

Table 2. The degree of nucleotide variation in three different loci (BenA, CaM, RPB2) within and between recognized new species, A. allahadadii and closely related clades.

^* n: number of isolates in each species/clade; ? RPB2 sequence data is available only for the ex-type isolate.

Taxonomy

Based on phylogenetic analyses of DNA sequence data, we identified A. flavus/A. oryzae (sect. Flavi), A. luchuensis, A. niger, A. tubingensis (sect. Nigri) and A. terreus (sect. Terrei) (Fig. 10) as known Aspergillus species together with a new species from section Terrei (ser. Nivei) named as A. mahabadiensis sp. nov., which is described and illustrated here.

Fig. 10. Colonies on MEA and micromorphology of identified Aspergillus species. (A-D) A. niger IRAN 5126C, (E-H) A. tubingensis IRAN 5122C, (I-L) A. luchuensis IRAN 5120C, (M-P) A. terreus IRAN 5116C, and (Q-T) A. flavus/A. oryzae IRAN 5119C. Scale Bars: B, F, J, N, R = 20 µm; C, D, G, H, K, L, O, P, S, T =10 µm.

Aspergillus mahabadiensis Abdollahz. & O. Ghaderi, sp. nov.; Fig. 11.

MycoBank: MB856047

Holotype: Iran, Western Azarbaijan Province, Mahabad, 36°45'11.3"N, 45°43'19.0"E, from Terkhêna a traditional food, October 04, 2021, O. Ghaderi (IRAN 18461F, living culture IRAN 4983C = CBS 151400).

Etymology: The specific epithet refers to Mahabad, a city located in the west of Iran from which the ex-type strain IRAN 4983C was isolated.

In: Aspergillus subgen. Circumdati sect. Terrei ser. Nivei.

Diagnosis: The species is phylogenetically distinct from other members of ser. Nivei. Phenotypically, it can be differentiated from all closely related species, i.e., A. allahabadii, A. barbosae, A. carneus and A. niveus, by faster growth rate on CZ at 25 °C (32–33 mm/1W), larger conidia (2.8–3 µm) and the presence of diminutive forms.

DNA barcodes: ITS = PQ846961; BenA = PP544444; CaM = PP544440; RPB2 = PP544442.

Colony diam, 7 d (in mm): CYA 25 ᵒC 40–41 mm; CYA 30 ᵒC 44–46 mm; CYA 37 ᵒC 52–53 mm; MEA 25 ᵒC 26–30 mm; YES 25 ᵒC 55–56 mm; CZ 25 ᵒC 32–33 mm; OA 25 ᵒC 24–25 mm.

Colony characters: CYA 25 ᵒC, 7 d: Colonies moderately deep, radially sulcate; margins entire, regular; mycelia white; texture floccose; sporulation moderately dense; conidia en masse rosy buff (61) to rosy vinaceous (58); sclerotia absent; soluble pigments absent; exudates clear or hyaline droplets; reverse pale luteous (11) to ochreous (44). YES 25 ᵒC, 7 d: Colonies moderately deep, randomly sulcate; margins entire, regular; mycelia white; texture floccose;  sporulation dense; conidia en masse in shades of white to pale vinaceous (85); sclerotia absent; soluble pigment absent; exudate absent; reverse pale luteous (11) to sienna (8) or bay (6). MEA 25 ᵒC, 7 d: Colonies moderately deep, radially sulcate; margins entire, regular;

Fig. 11. Aspergillus mahabadiensis IRAN 4983C: (A) 7 d old colonies at 25 ᵒC: top row left to right, obverse MEA, CYA, YES, and OA; bottom row left to right, reverse MEA, CYA, YES, and OA, (B) exudates on CYA, (C) exudates on CZ, (D) immature conidial head, (E) asymmetric foot cell, (F) conidia, (G) common biseriate conidial head and conidiophore, (H–I) proliferant vesicles (black arrows), (J, K) diminutive penicillate conidiophores, (L, M) reduced forms: (L) sessile metulae bearing phialides, (M) phialide directly on hyphae, (N, O) accessory conidia. Scale bars: D, E, G-K = 10 µm; F, L-O = 5 µm.

mycelia white; texture floccose; sporulation moderately sparse; conidia en masse in shades of white; sclerotia absent; soluble pigments absent; exudates absent; reverse white or hyaline. CZ 25 ᵒC, 7 d: Colonies moderately deep, plane, slightly concentric at the center; margins entire, regular; mycelia white; texture floccose; sporulation dense; conidia en masse in shades of white to pale vinaceous (85); sclerotia absent; soluble pigments absent; exudates clear to buff (45); reverse pale luteous (11) to sienna (8) or bay (6). OA 25 ᵒC, 7 d: Colonies moderately deep, plane, concentric; margins entire, regular; mycelia white; texture floccose; sporulation dense; conidia en masse in shades of white to pale vinaceous (85) or yellow (as small spots) at the center and white at the margin; sclerotia absent; soluble pigments absent; exudates absent; reverse white or hyaline.

Micromorphology: Conidial heads hyaline, radiate, biseriate. Conidiophores with smooth stipes, hyaline, aseptate, 164–301 × 2.3–7.6 µm, foot cell asymmetric. Vesicles hyaline, subglobose to subclavate, 6.2–18.9 µm wide, fertile over the upper half to two-thirds, rarely proliferant. Metulae hyaline, oblong to wedge-shaped, 4.4–8.2 × 1.5–3.1 µm. Phialides hyaline, flask-shaped, 3–5 on each metulae, 4.9–8.1 × 1.8–2.5 µm. Conidia hyaline, globose, smooth, (2.4–)2.8–3(–3.2) µm (2.9 ± 0.2, n = 50). Diminutive forms (small uniseriate or biseriate vesicles appearing penicillate, sessile metulae bearing phialides, phialides directly on hyphae) were observed. Accessory conidia hyaline, ovoid to ellipsoid, base truncate, 3.4–4.9 × 2.6–4.1 µm.

Additional specimens examined: IRAN, West Azarbaijan Province, Mahabad (36°45'44.6"N, 45°44'04.0"E), from dried apricot, 23 October 2021, O. Ghaderi, IRAN 4982C = CBS 151399; Miriseh village (36°29'51.8"N, 45°34'01.1"E), from dried apricot, 03 October 2021, O. Ghaderi, CJA OGhZ17 = DTO519-H3 = CBS 152878; Laj village (36°54'07.0"N, 45°45'55.0"E), from dried apricot, 23 October 2021, O. Ghaderi, CJA OGhZ61 = DTO519-H4 = CBS 152879; Yusef Kandi village (36°48'56.5"N, 45°44'09.9"E), from dried apricot, 25 October 2021, O. Ghaderi, CJA OGhZ75 = DTO519-H5 = CBS 152880; Jandaran village (36°24'59.3"N, 45°31'02.9"E), from Terkhêna a traditional food, 03 October 2021, O. Ghaderi, CJA OGhD15 = DTO519-H6 = CBS 152878.

Notes: Phylogenetic analyses revealed that Aspergillus mahabadiensis is a distinct species placed in ser. Nivei close to A. allahabadii, A. barbosae, A. carneus and A. niveus and some potential new species. Morphological examinations indicated that it is possible to separate A. mahabadiensis from closely related species. We have listed some important colony and morphological characters in Aspergillus taxonomy to differentiate A. mahabadiensis from closely related species (Table 3).

Table 3. Phenotypical features to distinguish A. mahabadiensis from closely related species in ser. Nivei.

DISCUSSION

People living in the Zagros region located in the west of Iran, traditionally process and store some foods needed throughout the year. Raisins and dried fruits such as apple, apricot and white mulberry together with Terkhêna, a popular dried traditional food, are the most commonly consumed foods. Food seldom seen as an ecosystem, possibly because it is not considered as a “natural” system. The association between foodstuffs and fungi has existed for thousands of years, consideredan important ecosystem (Pitt and Hocking 2009). Foods and dried fruits are subjected to fungal contaminations during process and storage (Wehner and Rabie 1970, Tournas et al. 2015, Snyder and Worobo 2018, Alghamdi et al. 2023). Fungi affect directly their quality and marketability, and on the other hand due to mycotoxin production indirectly threaten human health over time (Drusch and Ragab 2003, Trucksess and Scott 2008, Jard et al. 2011, Hartwig et al. 2020, Nikolchina and Rodrigues 2021, Naeem et al. 2022). A broad spectrum of fungi encompassing the most important genera Aspergillus, Fusarium and Penicillium are found in association with food spoilage and mycotoxin contaminations during processing and storage (Reddy et al. 2010, Houbraken and Samson 2017, Schaarschmidt and Fauhl-Hassek 2018). Aspergillus is a species-rich fungal genus encompassing the most important mycotoxigenic species, which are of significance in agriculture and food production and consequently for human, livestock and poultry health (Bhat et al. 2010, Jard et al. 2011, Kumar et al. 2016, Taniwaki et al. 2018, Ráduly et al. 2020). The precise identification of food spoilage fungi is a fundamental step to developing prevention and intervention strategies to minimize food waste and safeguard food quality (Filtenborg et al. 1996, Houbraken and Samson 2017).

As deduced from the literature, among the agricultural commodities few studies have considered mycotoxigenic fungi associated with dried fruits. Likewise, few studies, mostly on raisins, have considered fungi associated with dried fruits in Iran (Sedaghati et al. 2011, Ghezel Sefloo et al. 2012, Khodaei et al. 2014, Mirabolfathy et al. 2014, Sarabi et al. 2015, Amirmijani et al. 2018, Mehraban et al. 2019).

In an extensive study, we focused on the identification of mycotoxigenic fungi associated with raisins, some dried fruits including apple, apricot and white mulberry and Terkhêna a traditional food common and popular in Zagros region of Iran. A large fungal collection morphologically belonging to four genera Alternaria, Aspergillus, Penicillium (Ghaderi and Abdollahzadeh 2024a) and Rhizopus (Ghaderi and Abdollahzadeh 2024b) were isolated. A majority of the fungal isolates (76.4%) belonged to the genus Aspergillus. Using ISSR technique, our isolates were grouped into 11 clusters and thus we significantly reduced the number of isolates from 317 to 11 for phylogenetic and micromorphological studies. The isolates selected based on ISSR fingerprinting profile were placed in six Aspergillus species which is due to the fact that even though some researchers applied the DNA fingerprinting patterns for discrimination at interspecies level (Alves et al. 2007, Abdollahzadeh and Zolfaghari 2014), actually these techniques (e.g. ISSR, RAPD, rep-PCR, RFLP and SSR) are used for genotyping of fungal species in population genetic studies to calculate the genetic variation at infraspecies level and thus they can detect the degree of diversity more than the interspecies level. In this study, four selected isolates as representatives of four different DNA banding profiles were placed in A. flavus/A. oryzae, A. luchuensis, A. mahabadiensis, A. terreus) which indicates the ability of this technique to discriminate at the species level. In addition, the placement of seven isolates with different DNA fingerprinting profiles in two Aspergillus species (A. tubingensis and A. niger) reflects the genetic diversity within these two species and indicates that screening of our isolates using ISSR technique most likely didn't lead to underestimation at the species level and loss of any species. Sequencing of large collections is not affordable for all researchers around the world and phenotypic characters are not reliable thus it is recommended to examine the efficiency of DNA fingerprinting techniques in the discrimination of fungal species in more fungal groups especially in the important fungi such as Aspergillus as an interesting fungus for researchers in various scientific fields.

In a polyphasic approach based on morphology, DNA fingerprinting profile and sequence data, we identified A. flavus/A. oryzae, A. luchuensis, A. niger, A. terreus, A. tubingensis, and A. mahabadiensis sp. nov. We isolated A. mahabadiensis sp. nov., only from dried apricot and Terkhêna. It is distinguished from close species including A. barbosae, A. carneus, A. niveus and A. allahabadii in all single and multigene phylogenetic analyses of the main DNA barcodes CaM, BenA and RPB2. Moreover, A. mahabadiensis can be distinguished from the close species based on a combination of micromorphological features, colony characters and growth rates on different culture media. Among these, the presence of diminutive forms, conidial dimension, and growth rate on CZ at 25 °C are the most useful and effective characters in species recognition. Moreover, color of conidia en masse can differentiate A. mahabadiensis (pale vinaceous/rosy buff to rosy vinaceous) from A. allahabadii (White to glaucous blue), A. barbosae (Gray to greenish), A. niveus (White), but it is somehow similar with A. carneus (Pale vinaceous fawn to vinaceous fawn). Growth rates on CYA at 25 °C and 37 °C and on YES at 25 °C, which are available only for A. allahabadii and A. barbosae, are discriminative and faster in A. mahabadiensis. Accessory conidia produced by A. mahabadiensis have been reported for A. carneus and A. niveus but not observed in A. allahabadii and A. barbosae.

Based on a multigene phylogeny, Visagie and Houbraken (2020) detected a large degree of variation within A. allahabadii and recognized some clades with potential as new species in series Nivei but they didn't introduce them due to the small number of isolates and geographic origin. In our study, one of those clades consisted of isolates CBS 142986 and CBS 142987 recognized as conspecific with our isolates identified as A. mahabadiensis sp. nov., and the other clades are close to this new species. Aspergillus mahabadiensis sp. nov. is obviously differentiated from these potential new species based on CaM and specifically RPB2 and multigene phylogenies.

In our survey, A. tubingensis (n = 155) and A. niger (n = 120) with 275 isolates out of 317 Aspergillus isolates were determined as the two most frequent species, both isolated from all examined foodstuffs. We didn’t isolate A. flavus/A. oryzae and A. terreus from raisins and A. luchuensis from dried apple. We found all identified Aspergillus species on dried apricot and Terkhêna, and this is the first research on fungi associated with Terkhêna. So far as we know, some identified species here are reported for the first time on dried apple (A. terreus, A. tubingensis), apricot (A. luchuensis, A. tubingensis) and white mulberry (A. luchuensis, A. terreus, A. tubingensis) around the globe. Based on our knowledge, A. luchuensis is reported for the first time on foods in Iran and no information is available in the country on Aspergillus species on dried apple, apricot and white mulberry. Considering the consequences of mycotoxigenic fungi for human health, metabolite profiling and detection of mycotoxins especially for A. mahabadiensis sp. nov., in culture and substrates is significant to investigate in future studies.

 ACKNOWLEDGMENT

This study was financially supported by the University of Kurdistan. O. Ghaderi was financially supported by the ″Iranian Mycological Society″. Mahshid Saedi for PCR and sequencing of the DTO isolates.